Applied Statistics

National Industrial Work Training Service
PRODUCTION MECHANICS
Applied Statistics
Code: 81030751
Operational Level Technician

INDEX
Introduction
Applied Statistics...............................................................................................................1
Introduction.........................................................................................................2
INTRODUCTION TO STATISTICS.................................................................8
2. Division of statistics....................................................................................8
2.1 Descriptive statistics................................................................................8
85+90+ 93+82+95 = 415 = 89 points.................................................................9
2.2 Inferential statistics..................................................................................9
Issues:...............................................................................................................11
Solution:............................................................................................................15
Solution:............................................................................................................16
Review exercises related to basic descriptive statistics....................................16
1. Data collection...........................................................................................18
A)Some procedures and methods for collecting data.......................................18
Disadvantages of interrogation.........................................................................19
Interrogation methods.......................................................................................19
Frequency for data collection...........................................................................20
2. Identification of variables.............................................................................21
Example:...........................................................................................................22
Solution:............................................................................................................23
PRACTICE N° 2...............................................................................................25
INTRODUCTION TO STATISTICS APPLIED TO PROCESSES FOR
QUALITY IMPROVEMENT..........................................................................26
Statistical Process Control: X and R Charts.....................................................27
Basic philosophy of quality..............................................................................32
1.1 Introduction............................................................................................32
1.2 Concept and evolution of quality...........................................................33
Concept:............................................................................................................33
Quality Evolution..............................................................................................34
1.3 Understanding the meaning of total quality...........................................35
1.4 Strategic importance of total quality......................................................37
1.5 Total quality as a new business management system............................38
1.6 Reflections for the reader............................................................................38
The improvement process towards total quality...............................................38
2.1 Basic principles for achieving total quality................................................38
3. The production process is throughout the organization............................39
5. The supplier is part of our process............................................................39
6. Internal supplier-customer chains are essential.........................................39
7. Quality is achieved by people and for people...........................................39
8. Establish the zero defect mentality............................................................39
2
9. The competitive advantage is in the reduction of errors and continuous
improvement.....................................................................................................40
10. The participation of everyone is essential (collective consciousness)...40
ll.Quality...........................................................................................................40
12. It requires a new culture.........................................................................40
2.2 Improvement Modalities........................................................................40
a) Continuous improvement..........................................................................40
b) Radical innovation or improvement..........................................................40
2.3 The control cycle for improvement.......................................................41
2.4 Activities to start a process towards total quality.......................................41
PHASE I: Decision making..............................................................................42
PHASE II: Stage preparation and promotion...................................................42
PHASE III: Implementation of improvement processes..................................42
PHASE IV: Consolidation and interfunctional optimization...........................43
QA.....................................................................................................................43
QA.....................................................................................................................44
1. Functions and principles of quality control...............................................44
1. "Control is not classification or selection"................................................45
2. “With control you cannot obtain quality; This is an inherent characteristic
of the product.".................................................................................................45
3. "The production team is responsible for quality and control"...................45
4. "Control does not solve manufacturing problems, it only gives reasons to
study them".......................................................................................................45
5. Decisions must be made based on real data."............................................45
6. "The data must be compatible and arranged in such a way that it allows
analysis."...........................................................................................................45
QA.....................................................................................................................47
Why produce quality products? (Fig. 2)...........................................................47
Introduction to standardization.........................................................................50
1. Standardization..........................................................................................51
General concepts...............................................................................................51
Definition according to ISO..............................................................................52
Technical rules..................................................................................................53
Types of technical standards.............................................................................53
APPLICATION LEVELS................................................................................54
ISO 9000 STANDARDS..................................................................................54
ISO 9000 objectives..........................................................................................55
ISO 9000 standards...........................................................................................55
STANDARD IS014000....................................................................................55
IS014000 Standards..........................................................................................56
Terms and definitions (14050 Vocabulary)......................................................56
3
Frequency distribution table.............................................................................57
1. Discrete Variable Data..............................................................................57
Solution:............................................................................................................57
Ni=n+n2+na + ... + n i.........................................................................................58
Interpretation:...................................................................................................60
Properties of frequencies..................................................................................62
Solution.............................................................................................................66
Continuous variable data..................................................................................68
Construction of class intervals..........................................................................69
Lower limit Upper limit.................................................................................70
Determination of absolute frequency................................................................75
Observations:....................................................................................................76
Y1-y'i=<y'i-1 , y'i]..............................................................................77
General rule for developing frequency distributions........................................77
and. — 93.5494.5 — 94.0...................................................................................79
Definition 20: Percent cumulative relative frequency at................................82
Solution to:........................................................................................................83
+68...................................................................................................................................97
PRACTICE N°3..............................................................................................103
Time elapsed from receipt of order to delivery (in days)...............................104
AVERAGE.....................................................................................................106
Observation.....................................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X n )-n x
= n x -n x = 0...........................................................................................106
2x - X|20...........................................................................................................107
Example..........................................................................................................107
Solution...........................................................................................................107
ge1/gn240_24.........................................................................................................107
Proposition (Konig)........................................................................................107
> 2(xx)2.....................................................................................................................114
Observation.....................................................................................................114
Abbreviated calculation..................................................................................115
The harmonic mean........................................................................................115
The mean square.............................................................................................121
THE MEDIAN................................................................................................122
Observation.....................................................................................................122
Observation.....................................................................................................122
FASHION.......................................................................................................124
Observation.....................................................................................................125
EXAMPLE.....................................................................................................126
4
PARTIAL EVALUATION I..........................................................................127
RANGE, VARIANCE, STANDARD DEVIATION.....................................144
MEASURES OF DISPERSION.....................................................................144
= (0,010)= 0,0320...................................................................................145
Interpretation and application.........................................................................145
Breakdown......................................................................................................146
Where:.............................................................................................................146
Example..........................................................................................................146
X=1 2x...........................................................................................................................147
a = 12(x-7)2......................................................................................................147
°\4...............................................................................................................................147
°s\4.........................................................................................................................147
PRACTICE N° 4.............................................................................................148
from $10, from...........................................................................................................148
FREQUENCY HISTOGRAM........................................................................149
C ¡ xn =ni(ocxh= h) C ¡ Ci............................................................149
Frequency polygons........................................................................................151
Warhead cumulative frequency polygons.......................................................152
Note: Percentage warhead..............................................................................152
Solution:..........................................................................................................155
Line graphs.....................................................................................................158
PRACTICE N° 5...............................................................................................159
1. Concept:...................................................................................................165
2. Application:.............................................................................................165
4. Distribution function...............................................................................166
5. Typing.....................................................................................................167
_x-u....................................................................................................................167
-...........................................................................................................................167
— —- z..............................................................................................................167
Z
= npq is N(0,1).......................................................................................................168
Characteristic of the standard (reduced, standard) normal distribution..........168
6. Table management, most frequent cases.................................................169
Example:.........................................................................................................171
Concept:..........................................................................................................173
How to interpret a cause-effect diagram:........................................................173
Examples of cause-effect................................................................................173
The seven tools of quality...............................................................................174
H1 - Cause-Effect Diagrams...........................................................................174
Exercise: Cause - Effect Diagram...................................................................180
DISPERSION DIAGRAM.............................................................................180
5
LINEAR CORRELATION............................................................................182
SPEARMAN rank correlation coefficient......................................................184
> 6 2 d........................................................................................................................185
Yule “Q” correlation coefficient.....................................................................185
-
138 -:994.....................................................................................................185
V(a+b)(c+d)(a+c)(b+d)....................................................................................186
Correlation and Causality...............................................................................186
SIMPLE LINEAR REGRESSION.................................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0.....................................................................188
5b........................................................................................................................188
2 and xi = b 2x2+ aZx..........................................................................188
YsYb(X- x)............................................................................................................188
Concept...........................................................................................................188
What is it?.......................................................................................................189
When it's used?...............................................................................................189
How is it used?................................................................................................190
Relationship with other tools..........................................................................191
Application example.......................................................................................191
PRACTICE N° 7.............................................................................................195
Exercise: Pareto Diagram...............................................................................195
Organized:.......................................................................................................195
Work stoppage................................................................................................195
Getting relative percentage.............................................................................195
Construction of the diagram and determination of “vital few”......................195
Concept:..........................................................................................................195
Control elements.............................................................................................196
2. Control sheet...............................................................................................196
Scatter plots....................................................................................................197
A control chart shows.....................................................................................198
Need for full participation..............................................................................198
PARTIAL EVALUATION II.........................................................................200
THE QUALITY ROUTE...............................................................................205
First Step: Defining the Problem....................................................................205
Second Step: Recognition of the Characteristics of the Problem (Observation)
Activities.........................................................................................................206
Third Step: Search for the Main Causes (Analysis) Activities.......................207
Fourth Step: Actions to eliminate the causes (Action) Activities...................207
Fifth Step: Confirmation of the effectiveness of the action (Verification).
Activities.........................................................................................................207
Step Six: Permanent elimination of the causes of the problem (Standardization)
........................................................................................................................207
6
Activities:........................................................................................................207
Step Seven: Review of activities and planning of future work (Conclusions).
........................................................................................................................208
Activities.........................................................................................................208
•
• Bibliography
7
SENATI APPLIED STATISTICS
INTRODUCTION TO STATISTICS
The word "statistics" like many other words has several meanings.
In common language, the term "statistics" means a little more than data or
numerical information and is generally used in the plural. For example, the
sports writer, while commenting on the incidents of the first stage of a soccer
match, can say, these are the statistics for the first stage; team A: shots on goal
ten, nine missed and one successful, corner kicks six, etc., team B: shots on
goal eight all missed, corner kicks seven, etc. Radio or television commentators
agree by saying: the statistics of the Gross Domestic Product (GDP) of the
Manufacturing Industry went from 15.3% in 1988 to 11.7% in 1990.
Between March 88 and August 90, salaries have fallen by 60%. According to
the National Institute of Statistics and Informatics (INEI), the accumulated level
of inflation during the first half of this year reached 230.5%. A random person
can ask: have you posted the latest statistics about employment?, etc.
Naturally, in each of these comments, people are using the word "statistics"
correctly, however, each uses it in a different way for a different purpose. The
term "statistics" is also used to designate an area of study. A discipline; for
example, when this word is used in the title of a book or when we refer to the
statistics courses in the curriculum of a profession, etc. However, both
meanings of "statistics" are closely related because "statistics" considered as
numerical data are largely the "raw material."
The evolution and development of "Statistics" in today's world is such that it

would be difficult to give a precise definition of this concept. At the risk of adding
one more to the many existing ones, we will give the following.
1. Definition: Statistics is a science that provides a set of methods used to

collect, summarize, classify, analyze and interpret the behavior of "data" with
respect to a characteristic subject of study or research. In the first instance, it is
responsible for obtaining information, describing it and then using this
information in order to predict "something" regarding the source of information.
Current Statistics is the result of the union of disciplines that evolved
independently until coming together in the 19th century: the first is the "calculus
of probabilities", which was born approximately in the 17th century as a
mathematical theory of games of chance; The second is "Statistics" or science
of the state, from the Latin status (although regarding this etymological meaning
of statistics there is no single criterion of the authors, since for some it is derived
from the Greek statera which means balance, for others it is derived from the
German staat meaning state) which studies the description of data and has
older roots. The integration of both lines of thought gives rise to this new
science.
2. Division of statistics
The field of Statistics is generally divided into two large areas.

Descriptive Statistics and Inferential Statistics.
2.1 Descriptive statistics

From the definition of Statistics, the reader will notice two well-
highlighted aspects, first: obtaining information, processing it and
PRODUCTION MECHANICS 10/14

8 7
describing it, is what constitutes Descriptive Statistics.

Descriptive Statistics is the set of methods that involves the collection,
presentation and characterization of a set of data in order to
appropriately describe its various characteristics. That is, a statistical
study is considered "descriptive" when only the data is analyzed and
described.
Example: A personnel manager wants to know the skills of five
secretaries who work in a particular department of a company. An
aptitude test is applied to the five secretaries and the scores are
85,90,93,82 and 95 points.
Suppose the statistical measure the personnel manager uses is
aptitude.
Average or arithmetic mean, which is the sum of the observed values

divided by the number of observations. So the average rating is:
85+90+ 93+82+95 = 415 = 89 points

5 5
The result is limited to the data obtained in this particular case and does
not imply any generalization about the skills of secretaries in other
offices of the same company. That is to say. The manager is using
statistics to describe the skills of the secretaries in that office. This
method is descriptive in nature, because the average condenses and
describes the information obtained.
Graphs, tables, and maps that display data in a way that makes it easier
to understand are all examples of using Descriptive Statistics.
2.2 Inferential statistics
The second aspect of the definition of Statistics is: predicting

"something" with respect to the source of information, which is what
constitutes Inferential Statistics or Statistical Inference, the definition
is as follows.
Statistical Inference is the set of methods or techniques that enable
generalization or decision-making based on partial information obtained
through descriptive techniques.
That is, a statistical study is considered inferential when it seeks to infer
and predict conclusions that concern the entire source of information
from which the data comes.

9 7
APPLIED STATISTICS
Now this prediction is made with a certain degree of confidence; This

degree of confidence is measured by "probability." Therefore, the
calculation of probabilities, the cornerstone of statistical inference, is a
bridge between the two parts of statistics.
Although descriptive statistics are important for characterizing and
presenting data information, however, the development of statistical
inference is what has led to the great expansion in the application of
statistical methods.
Example 1: Now suppose in example 1, that the personnel manager
wants.
Know the average aptitude of all the company's secretaries, but lack the
time and resources to apply an aptitude test to all of them. You then
decide to use the average aptitude of the five secretaries to estimate the
average aptitude of all the company's secretaries. The process of
estimating this global average aptitude will be a Statistical Inference
problem.
Example 2: A drug manufacturer claims that a new cold vaccine
developed by its company is 95% effective, that is, on average 95 out of
every 100 people who use the vaccine will make it through the winter
without catching a cold.
Since it is impossible to test the vaccine on all people, let us consider

that 40 people have received the vaccine, of which 40.35 did not catch
the cold. We see that if the manufacturer's statement is correct, 38
people (40 x 0.95=38) would be expected to spend the winter without a
cold. Since the observed number is 35, which is less than the expected
number 38, should the manufacturer's claim be rejected based on the
evidence? The decision process of whether or not to reject the
manufacturer's claim is a Statistical Inference problem.
The problem can also be stated as follows: The proportion of non-

infected people, which is the quotient of the number of observed people
who were not infected among the total number of observed people. That
is, the proportion of people not infected is:
35 7 = 0-88
40 8
This proportion expressed as a percentage is 100(0.88)% = 88%. So the
evidence indicates only 88% effectiveness of the vaccine in question, which is
lower than the manufacturer's estimate. The process of deciding whether or not
to reject the manufacturer's claim is a problem of statistical inference.
Example 3: Four brand A bulbs stopped working after 1100, 980,900 and 1020
hours of continuous use. Five brand B bulbs stopped working after
960,1050,1065,845 and 980 hours of continuous use. The following conclusions
are reached:
a. The average duration of the four brand A bulbs is 1000 hours, while the
average duration of the five brand B bulbs is 980 hours.
b. The average life of all brand A bulbs is longer than that of all brand B bulbs.

1 7
0
c. The difference between the two averages is 20 hours.
d. The difference between the two averages is too small to conclude that brand
A bulbs are better than brand B bulbs.
e. If another brand A bulb is selected and tested, it will probably last longer than
the average brand B bulb.
f. You decide to buy brand A bulbs instead of brand B bulbs.
Which of the conclusions come from Descriptive Statistics and which from
Statistical Inference?
a. The average duration of the four Aes brand bulbs
1100 + 900 + 980 + 1020 _ 4000 _ .000 is

4 4 .
The average life of the five brand B bulbs is
1050 + 960 + 1065 + 845 + 980 _ 4900 _ ho

980 hours
4 4 .
Since Statistics is being used to describe the behavior only of the observed data
and not for a generalization, the conclusion comes from Descriptive Statistics.
b. Since it is generalizing that the average lifespan of all brand A bulbs is longer
than all brand B bulbs, the conclusion comes from Inferential Statistics.
c. We see that: 1000 - 980 = 20 hours, that is, the conclusion comes from
Descriptive Statistics.
d. The averages of the observed data are being used to infer about the
difference of all the bulbs. Therefore, it is Statistical Inference.
e. A generalization is made that any brand A bulb will last longer than the
average brand B bulb, comes from Statistical Inference.
f. You are generalizing that brand A bulbs last longer than brand B bulbs.
Therefore, it is Statistical Inference.
Issues:
In an aptitude test, four workers received scores of 85,90,82, and 83. Four
workers received scores of 88,87,89,92. Of the following statements made
based on these qualifications, identify those that are derived from descriptive
methods and those that are derived from Statistical Inference.
a. The average rating of the four male workers is 85, and the average rating of
the four female workers is 89.
b. The average aptitude of all female workers is probably higher than that of
male workers.
c. On the next skills test, male workers will probably receive lower scores than

1 7
1
female workers.
2. Five automobile tires of brand A and four tires of brand B are tested for their
service life. The duration for brand A tires are: 36,000, 29,000, 33,000, 37,000
and 40,000 km; for brand B, they are: 29000,31000,33000 and 35,000 km.
Of the following statements made based on these figures, identify those that
come from descriptive methods and those that come from statistical inference.
a. The average life of the five brand A tires is greater than that of the four brand
B tires.
b. Probably, the average life of all brand A tires is almost 35000 km, while that
of brand B tires is approximately 32000 km.
c. If the price of brand A tires is the same as that of brand B tires, you would
recommend brand A tires to all your friends and relatives.
3. On four math tests, Juana received scores of 17, 18, 15 and 14 while Juan
received scores of 14, 13, 16 and 13. From these data the following conclusions
are reached, which of these include the descriptive method and which are
obtained through statistical inference?
a. Juana's average grade is 16 and Juan's average grade is 14.
b. Juana is a better student than Juan.
c. Probably on the next test, Juana will obtain higher grades than Juan.
d. The difference between the two averages is 2 points.
4. In what context is the word "statistics" generally used on radio than on

television?
5. Classify each of the following statements as either inferences or descriptive

methods.
a. Last year, at the National University of Callao, the average admission exam
score was 85.
b. Dr. García, an ecologist, reported that in a certain jungle river the fish meat
contains an average of 300 units of mercury.
c. The company "RM" predicted who would be the winner in a presidential

election after knowing the voting results of 25 polling stations out of the 2,800
polling stations in total.
3. Population and sample
The word "population" as with the term "statistics" has several meanings. In
common use, it refers to all people in a region, locality or country. (Lima's
population is 6 million; China's population is 900 million, etc.).
The concept of "population", statistically speaking, is a broader concept than

what is available when referring to the inhabitants of a Department or a Country;

1 7
2
Population from a statistical point of view is defined as follows.
Population: It is the collection of all individuals, objects u

observations that have at least one common characteristic.
The terms population and universe are often used interchangeably.

Example 4
a. The ages of students in Peru.

b. The ages of the students of the Peruvian university system.
c. The diameters of the daily production of nuts.
d. The weights of the melons from a harvest from the "meloneros" cooperative.
e. The license plates of cars that circulate in a country.
f. Patients with poliomyelitis undergoing a specific rehabilitation therapy.
g. AIDS sufferers, treated with one of three different treatments.
In each of the previous examples, there is at least one common characteristic

for all the elements that make up the population. And each different situation
involves a different population. In the 1.5 g example, instead of one population,
we have three different populations, each population is defined by each of the
three treatments used.
It is important to define the population according to the nature and extent of the
problem under study. When talking about nature, we will understand the
characteristic subject of study; That is, if we wanted to study the weights,
heights and ages of people, the population will be made up of the weights,
heights and ages respectively. And when talking about extension we define the
population as extensive as necessary; That is, if we wanted to study the above
characteristics in students of the Peruvian university system, the population will
refer to the Peruvian university system; and if we were interested in studying
these characteristics in the students of the universities of Lima, the population
will refer only to them.
When referring to the nature of the problem, we also choose to distinguish

between what we can call the "object" population and the "target" population;
We understand by target population, the set of elements subject to study and by
target population the different measures of the characteristic of the target
population that interests us. Let us clarify this with the following example: when
studying the academic performance of students in the university system; The
target population will be made up of the system's students and the target
population will be made up of the grades, which measure academic
performance. It is also necessary to note that:
(a) . Every population must be defined in such a way that the location of any
element in it can be decided unequivocally.
(b) . Any population can be real or virtual.

On the other hand, note, for example, when talking about "students of the
university system", we define a limited, therefore finite, population. On the other
hand, when we talk about "patients with poliomyelitis, subjected to a certain
rehabilitation therapy" we define an unlimited population, therefore virtually
infinite, whose elements are both the people who have already been subjected

1 7
3
to the aforementioned therapy, as well as those who be in the future. That is,
the population can be classified as finite or infinite.
Finite population: A finite population is one that has a limited number of
elements. For example, the heights of all the students currently studying at
universities in Peru.
Notation: We will denote the size of the finite population by "N".
Infinite population: An infinite population is one that has no limit; bounds, that
is, it has an infinite number of elements. For example, the quality of all units
produced through a manufacturing process.
Parameter: It is a summary measure that describes a characteristic of the

entire population.
Parameters are measurable characteristics of a population and naturally to
determine their value it is necessary to use information from the entire
population.
For example, the average age of students in the first year of secondary school
in Peruvian schools is a measurable characteristic (therefore a parameter) of
the population formed by the ages of all students in the first year of secondary
school in the country; It is therefore the average of the population.
Similarly, the proportion of all viewers who watch a certain Sunday program at a
certain time is a measurable characteristic (hence a parameter) of the
population made up of all viewers of that Sunday program; is the proportion of
the population. The most used parameters are:
- The population mean = u (read mn)

- Population proportion = p (read pe)
- Population standard deviation = o (read sigma)
Obviously it is impossible to calculate the true value of any parameter of an

infinite population. In most cases it is also not practical (or economical) to
calculate the true value of any parameter of a finite population. Therefore, it is
necessary to make inferences about the population parameters, based on the
information contained in a part of the population.
Sample: It is a representative part or subset of the population. And the process
of obtaining the sample is called sampling.
Notation: The number of observations (or size) in the sample is denoted by "n".
Of course, the number of observations in a sample is less than the number of

possible observations in the population, otherwise the sample would be the
population itself.
The selection and study of a sample aims to extract conclusions that are valid
for the population from which said sample was obtained. In other words, our
purpose is to know the population, for which a sample is extracted from it.
For example, if you wanted to estimate the average annual expenditure of

university students in Peru, you would extract a sample made up of a certain

1 7
4
number of students, then determine the annual expenditure corresponding to

each of them and then obtain the average These expenses.
A sample is used because there is simply not the time and resources to contact
all university students in the country, even though it is possible to do so.
Based on the sample average (sample average) obtained in this way, an

inference is made about the average expenditure of all university students in
Peru.
We have also expressed that it is impossible to calculate the true value of any
parameter of an infinite population, it is therefore necessary to use sample
information. For example, to determine the proportion of defective parts
produced in a certain manufacturing process, quality control technicians
examine a batch of produced units to determine the number of defectives
contained in 1. (Generally such a batch, which constitutes a sample, is taken at
regular time intervals).
The population proportion, which is an unknown parameter, is the proportion of

all defective units produced in the process; It is estimated by the sample
proportion, which is the proportion of defective units contained in the sample.
Thus, if in a batch of 300 defective units produced in the process, the quality
control engineer finds 45 defective, then the proportion of defectives in the
sample will be 45/300 = 0.15(15%).
Note that both the sample average and the sample proportion in the examples
considered in the previous paragraph are measurable characteristics of the
samples, which gives rise to the following definition.
Statistician or Statistician is a summary measure that describes a
characteristic of the sample.
For example, Cl (Intelligence quotient) average of a sample of first-grade

students selected from all first-grade students in Peru, is a statistic; Well, it is a
characteristic of the sample. Similarly, the proportion of a sample of viewers of a
certain program at a certain time is also a statistic; since it is a characteristic of
the sample of television viewers.
Example: Suppose that the entrants to the first year of your university consist of
3,000 students, all of whom have taken a single selection exam that was
applied to all students who have entered the first year of university in the
country. Explain the circumstances under which the grades received by
students entering the first year of your university can be considered as: a. a
sample, b. a population.
Solution:
a. There may be more than one circumstance. For example, if you want to
know the average grade of all the entrants to the universities in the country,
the population would be made up of the grades of all the entrants to the first
year of university in the country, then the grades of the entrants to your
university It would be a sample.
b. There may also be several circumstances. For example, if you want to know

1 7
5
the average grade of all entrants to your university. The population would be
made up of the grades of all those entering your university. We will thus have a
circumstance in which the grades received by students entering the first year of
their university are considered as a population.
Example: A preliminary vote is taken to determine the preferences of voters in

a presidential election. For this purpose, 1,500 registered voters are interviewed
and among them 860 are in favor of candidate A. Answer the following:
a. What constitutes the sample?

b. What constitutes population?
c. Is the population finite or infinite?
d. What is the population parameter?
e. What is the sample statistic?
Solution:
a. The sample consists of the responses of 1,500 registered voters.
b. The population is made up of the responses of all registered voters.

c. The population is finite.
d. The population parameter will be the proportion of all registered voters

who are in favor of candidate A.
e. The sample statistic is the proportion of the 1,500 registered voters who
favor candidate A.
That is, 860/1 500 « 0.57 (57%).
PRACTICE N° 01
Review exercises related to basic descriptive statistics
1. During a certain week, a restaurant served 2,300 customers. Explain the

circumstances under which these 2,300 customers can be considered: a. a
sample and b. a population.
2. Suppose that 60% of all registered voters in a country are members of party
A and 40% are not. From a sample of 500 voters, it is found that 250 belong
to party A. Answer the following:
a. What is the proportion of voters in the sample who belong to party A?

b. What is the proportion of voters in the population who belong to party A?
c. What is the population? Is it finite or infinite?

1 7
6
3. Establish the truth or falsehood of each of the following propositions. Replace

each false statement with the corresponding true statement.
a. The population is a collection of all the elements we are studying.

b. A statistic is a characteristic of the population.
c. Statistical Inference formulates inferences with respect to a sample.
4. An opinion sampling is carried out to determine if the housewives of Lima

prefer a detergent from brand A over another. For this purpose, 2,000
housewives are interviewed and among them 1,500 prefer brand A. Answer
the following:
a. What constitutes the sample?

b. What constitutes the population?
c. Is the population finite or infinite?
5. For each of the following statements, define the population being sampled
and describe the population parameter and the sample statistic.
a. 800 university students are interviewed about their opinions (yes or no

answers) on the legalization of abortion.
b. 300 non-union workers are interviewed to determine their opinions (yes or

no answers) about unionization.
c. A sample of the thymes produced by an automatic machine is selected

and tested to determine the proportion of defectives.
d. A sample of calves born in Cajamarca in 1989 is selected and a record of
their birth weights is kept.
6. Define appropriate populations and from them select the following samples:
a. People from 200 households in a certain city were called by telephone and
asked to name their candidate for mayor.
b. 200 pairs of a new type of shoes for soccer players were tested in a
professional tournament and. On average they lasted 4 months.
c. On five different occasions, it took an Engineer 21,26,24,22 and 21

minutes to drive his car from his home on the outskirts of the city to his
office in the city center.
VARIABLES AND DATA COLLECTION

1 7
7
1. Data collection
A)Some procedures and methods for collecting data

There are three basic methods by which the researcher can obtain the
desired data: first, the researcher can resort to data already published by
government, industry or individual sources; Secondly, you can design an
experiment to obtain the necessary data, and thirdly, you can conduct a
survey.
Sources of information: It is the place, the institution, the person where the
data needed for each of the variables or aspects of the research is. Data
sources can be:
Internal data source: It is the information collected by the company (or

institution) from the results of its own management. They are therefore the
observations that are constantly made by administrative, accounting,
commercial, technical, etc. departments. These can be for example:
a. Financial reports.
b. Operations reports, which are given by information on production, sales,
purchases, profit and loss statements.
c. Special reports are additional information for specific analysis.
a. External data source: These are statistical information prepared by

research institutions, whether public or private, or specialized agencies,
generally required at the national or sectoral level.
b. Primary sources: When the statistical information is obtained directly from
the observation unit. For example, the results of population and housing
censuses, consumer price indices, etc.
c. Secondary sources: When statistical information prepared based on data
from primary sources is obtained.
The fundamental official body for statistical data is the National Institute of
Statistics and Informatics (INEI). It is responsible for formulating and developing
the National Statistical system, carrying out national population and housing
censuses every 10 years and economic censuses every ten years, etc.
The Ministry of Tourism and Integration Industry prepares statistics such as:
physical production volume index, gross domestic product, etc.
The Ministry of Economy and Finance, through its specialized agencies,
maintains and prepares statistics on wholesale and retail trade, foreign trade
(import and export), etc.
In general, all ministries have a specialized Statistics office.
Experimental design: A second method of collecting information is with an

experimental design. Experimental designs should be used whenever possible
when you want to build explanatory models, and their studies are usually the
subject of more advanced texts, since they involve complex statistical
procedures.
Surveys: It is the process of collecting information through a sample.

Main procedures for collecting information in social sciences, health,

1 7
8
education, etc.
All information has two fundamental aspects: Source of obtaining, and

methods for its collection.
Source of obtaining: The source of obtaining can be: primary and secondary.
Primary: As we have already said, they are collected directly from their origin.
Secondary: When they are not collected directly from their source of origin.
Collection methods: When a primary source is used to collect information, two

fundamental procedures are distinguished, namely: observation and
interrogation.
Disadvantages of observation
- Requires specialized personnel.

- It may prove to be too expensive a method.
- It is not convenient when studying large human masses.
- An issue that invalidates observation is when research is required.
- The subjective manifestations of individuals. For example, knowing if a
patient has an appetite.
- In his past behavior. For example, illnesses you have suffered. Their future
attitudes. For example, what hygiene habits you will observe when you are
discharged.
Advantages of interrogation: As can be seen, the limitations of observation
are advantages of interrogation, since when the past or future is investigated,
as well as the subjective conditions of the individual, it offers satisfactory
results.
Disadvantages of interrogation
- An appeal is made to the memory and/or good faith of the person being
questioned.
- It produces different results depending on the type of questions and the
manner of forms.
To avoid errors in the second disadvantage, the following aspects must be

observed:
- Questions should be clear and concise.

- Questions should not be leading or ambiguous.
- Questions should not presuppose facts.
- Questions should not suggest answers.
Interrogation methods
The interrogation can be done by two methods:
Direct method (carried out through interviews)

Indirect method (carried out through questionnaires)
The interview has a main advantage that accentuates it, and that is that it can
be completed with direct observation. It also has as its main disadvantage the

1 7
9
fact that the personality, social position, the inflection of the voice, the way of
asking questions, etc. They can vary the answers.
The main advantage of the questionnaire is that it is cheaper. As a

disadvantage, it is only possible for simple questions and sometimes it is not
returned by the individual surveyed, or it does not fill all the requirements
adequately.
Frequency for data collection
The procedures for collecting information, due to their frequency, can be

classified as:
- Occasional collection methods: surveys.

- Periodic collection methods: censuses.
- Continuous collection methods: records.
These methods differ from each other in a series of important aspects, which
are:
- Collection frequency.
- Coverage aspects.
- The aspects of temporality.
- The aspects of purposes.
We will see that the survey has a partial amplitude, while the census and the
registry have a universal amplitude. Another very important aspect is that the
census and the survey have a transversal character, that is, they are carried out
at a certain moment, while the registry has a longitudinal character, that is, the
information is collected over time.
Finally we will say that the census is a procedure that has general objectives,
while the registry and the survey have specific objectives.
Survey: It is the procedure for obtaining information structured according to

previous systematization criteria, which is carried out for a specific purpose (and
which has restricted scope) in a sector of the population.
Example: National multiple-purpose survey (ENAPROM) carried out by the INEI

in 1977 and 1978, and published in 13 fascicles by this institution.
Types of surveys: Retrospective survey, prospective survey.
Retrospective survey: In this type of survey, the data is known and the
research consists of discovering characteristics of its history. For example, a
group of people affected with lung cancer is taken and we are going to
retrospectively record their history if they have a history of: smoking, industrial
risks, etc.
Prospective survey: Begins with a sample of the population studying one or

more characteristics over time. For example, we take a group of people, and we
look at the characteristic of their smoking habit: we see if lung cancer develops

2 7
0
over time.
Population census: The process of collecting, completing and publishing

demographic, economic and social data belonging to a specific time and data
on all people in a country or a given territory.
The main purpose of a national population census is to satisfy certain statistical

information needs concerning all the inhabitants of a country with a view to:
planning a series of aspects such as food requirements, the numbers of schools
and hospitals , as well as its location, etc.
Records: These are the procedures that are followed to adequately know the
changes and structures of:
- Changes in the number of the population.

- Increase in population by birth.
- Increase in population due to immigration.
- Reduction in population due to emigration.
- Reduction in the population due to deaths.
Through registration systems, a continuous process is established that without

pause follows the movements of change in the population.
2. Identification of variables
Whatever the source from which we obtain the information, it may refer to
qualitative or quantitative characteristics. The first refer to qualities such as
color: white, blue, etc.; marital status: married, single, etc.; profession:
economist, engineer, etc.; quality of a product: good, average, etc. The latter
refer to quantities such as: height in cm, salary in soles, number of children in
a family, number of bedrooms per home, etc.
With each characteristic under investigation, we associate a variable; Thus,

we can give the following definition.
Variable: It is a characteristic of the population that is going to be investigated

and that can take different values.
Thus, for example, a variable would be the overtime hours worked by the
workers of a company and the values of this variable would be given by the
different hours worked by each worker outside the normal working day: none,
one, two, three,
Notation: Variables will be denoted by X, Y, etc.
The variables are classified as: qualitative and quantitative.

Qualitative variable: This is called when the variable is associated with a
qualitative characteristic. That is, they are variables whose values are qualities
that the population presents.
For example, the variable "profession" can take the following forms: Engineer,

2 7
1
Doctor, Biologist, Economist, etc.
The qualitative variables are classified as: nominal and ordinal.

Nominal qualitative variable: These are those that establish the distinction of
the elements in the categories without implying order between them.
Example: Classify a group of individuals by sex: male, female; by their marital

status: single, married, widowed, etc.
Ordinal qualitative variable: These are those that group objects, individuals,
into ordered categories, to establish comparative relationships. That is, they are
susceptible to organization but not quantitative measurement.
Example:
a. Classify a group of people by their smoking habit. It proceeds like this:

- Non smoking
- Light smokers
- Moderate smokers
- Severe smokers
b. Classify a group of individuals by their level of education. Proceed as

follows:
-Illiterate
- Primary
-Secondary
-Superior
Quantitative variable: This is called when the variable is associated with a

quantitative characteristic. That is, these arise when it is possible to establish
how much or in what quantity a certain characteristic is possessed.
Example: Quantitative variables are: family income, number of traffic accidents,

length, time, etc.
Quantitative variables are divided into discrete and continuous.
Discrete variable: These are those that arise from the counting procedure.
That is, discrete statistical variables usually take integer values.
For example, the number of children per family; the number of students per
school; the number of cars that pass along an avenue in an hour; the number of
inhabitants per district; etc., are discrete variables.
Continuous variable: These are those that arise when some characteristic is
measured. That is, continuous variables can take, at least theoretically, any
value within an interval.
For example, people's weight, height, blood pressure, income, length of service,
etc., are continuous statistical variables.
It does not mean that the number assigned to each weight, height, etc., can in
practice take any real value of an interval (although theoretically this is the
case) because the imprecision of our measurements will ultimately mean that
they are represented by discrete values. But it is the character of possibility that

2 7
2
every real number has of being assigned to one of these variables that defines
them as continuous.
In general, all quantities related to time (age, duration of a phenomenon), mass

(volume, weight), space (length, surface) or a combination of these (speed,
density, capacity) are variables. you continue.
Example: Classify the following characteristics into variables
qualitative, continuous, discrete.
a. Length of service of a company's employees.
b. Number of checks drawn daily in a month.
c. Number of common shares sold each day on the Lima Stock Exchange.
d. Birthplace of people who live in Lima.
e. Educational level and religion of Lima women.
f. Daily temperature and humidity of Lima.
g. Educational level, height and eye color of Cajamarcans.
Solution:
a. The time that any person provides services to a company is generally

included according to certain laws between 0 and 30 years. That is, it can be
any point in the interval [0, 30], at least in theory. Therefore it is a continuous
quantitative variable.
b. The number of checks drawn daily in a month can be: none, one, two. That
is, it is a discrete quantitative variable.
c. It is also a discrete quantitative variable.
d. People who live in Lima. They may have been born in any district or place in
the country or abroad. Therefore, it is a qualitative variable.
e. Educational level is an ordinal qualitative characteristic and religion is

nominal qualitative. Therefore, Lima women are classified by the qualitative
variables, educational level and religion.
f. Temperature and humidity are continuous quantitative variables. Then, Lima

is classified by the variables temperature and humidity will be a continuous
variable.
g. Educational level and eye color are qualitative variables; Height is a

continuous quantitative variable. Therefore, Cajamarcans are classified by
qualitative characteristics: educational level and eye color, and the
continuous variable height.
Statistical unit: It is the indivisible element or object of the population that will
be analyzed.
For example, if you want to study the demand for detergent X in Metropolitan

2 7
3
Lima, the statistical unit will be families.
Data: These are the values collected as a result of observations of a

characteristic or variable. Also called statistical series.
As there are two types of variables: qualitative and quantitative, we will also say
that there are two types of data: qualitative and quantitative data. Quantitative
data can also be considered as discrete data and continuous data.

2 7
4
PRACTICE N° 2
1. Can you sort people by their marital status: single, married, widowed,
divorced? Justify your answer.
2. In a horse race, Saltarín came first, Pintado, second, and Ronco, third. Can
you find the distance between the horses? Justify your answer.
3. Classify the following characteristics into qualitative, quantitative continuous

and discrete variables.
a. Monthly current consumption (kw) in a year.

b. The number of screws in each box produced in a month.
c. Marital status, public opinion, place of birth of people who live in Callao.
d. Brand and country of origin of automobiles sold during the year in Peru.
4. If the only possible values for a variable 1/3, 1/4,^,

so X is a continuous variable.
5. A customer warranty policy is included with every device produced by an

electrical equipment company. In addition to validating the guarantee and
providing the company with the customer's name and address, the policy
requests other complementary information that is used in market studies. For
each of the numbered policy targets, determine the most likely characteristics
of the categories the company will use to record information. In particular:
a. Will they be quantitative or qualitative?

b. Continuous or discrete?
Name __________________ __ Civil status ________________________

Home __________________ Where was the equipment purchased?
City State •
Why was the equipment purchased?
Postal Code______________
Age _OR_ Annual Income _ (2) ©
6. Which of the following statements represents the most accurate scheme for
classifying data?
a. Quantitative methods.
b. Qualitative methods.
c. A combination of quantitative and qualitative methods.
d. The scheme is determined only by concrete information about the
situation.

2
5
7. For each of the following statements, identify the statistical unit:
a. The mayor of a city wants to estimate the percentage of voters in his

district who will favor the passage of a municipal law.
b. A supermarket chain wants to know its employees' opinions about a

company-sponsored health insurance plan.
c. A company marketing executive wants to survey buyers to determine their

attitude toward a new product line.
8. In the following statements, identify: the population, the sample, the variable,
the type of variable, the statistical unit, the parameter and give an
observation.
a. A study is going to be carried out on the amount of sugar sold per week, in
a supermarket in a certain sector of Lima, for the year 1990; for which the
average sales of that year will be used, the records are made in kgs. per
week.
b. An industrial firm, engaged in the production of steel, produces screws for

which there are narrow tolerance margins with respect to width, a number
of screws are selected from the daily production and carefully measured.
INTRODUCTION TO STATISTICS APPLIED TO PROCESSES FOR QUALITY

IMPROVEMENT
In Taylor's day, all plants, with a few exceptions, had an inspection department.
Its function was to eliminate unacceptable materials from the production
process, from the input material to the finished product. At that time, statistical
sampling was eminently unknown in manufacturing, so key points in the
process were frequently chosen at which 100% of the inspection was carried
out.
Because employees were often paid based on the number of good units
produced and production managers were evaluated based on the number of
units shipped, inspection department employees and managers traditionally
lived in environments that were often poor. friendly and sometimes hostile.
Quality control as a subdiscipline of industrial engineering has two basic

components. One component involves the technical use of applied statistics
and is known as statistical quality control (SQC) or statistical process control
(SPC). The other component involves the development and day-to-day
administration of a programmatic and administrative basis for achieving quality
control objectives in an organization.
As stated before, no course in a typical industrial engineering curriculum is

more important than a course in probability and statistics. This is especially true
in the area of statistical quality control. Statistical quality control, in terms of
technical content, is simply the application of statistics and sampling theory in

2 7
6
an industrial or productive environment.
With all the importance of probability and statistics to the education of an

industrial engineer, no attempt will be made in this text to describe the content
of such courses. We assume that a course in statistics is an essential
requirement for the study of statistical quality control.
Statistical Process Control: X and R Charts
Probably the best known, and perhaps the oldest, technique in statistical quality
control (SQC) is the X and R chart. It was originally developed by Shewhart [14]
in the 1920s. The graph is used when the goal is to maintain statistical control
of an isolated variable of interest (for example, a dimension for a component or
assembly).
Figure 1 describes a cutting operation on a lathe, with dimension A of interest.

The operation could be performed, for example, in preparation for cutting
threads to join two pipes in an oil field (e.g., in West Texas).

2 7
7
Tool movement wh
External diameter Initial rod

at desired end | Screwing diameter
machine
Field
measurements,
to the nearest
W
hundredth of an
Fig. 1 Final external diameter A on
inch, for the first Here the
screwing machine.
piece is cut
48 pipe cuts are
given in Table 1.
Note that the measurements have been divided into samples of size four. As
explained elsewhere in this book, the reason for doing this is to take advantage
of the central limit theorem, which basically says that when you graph the
means of samples from an unknown but essentially unimodal distribution of
individual values, the means will have a relatively normal distribution, that is, a
Gaussian distribution. (It is not true, as a graduate mathematics student stated
during an oral exam after failing to remember the word Gauss, that the original
creator of the normal distribution was called Sir George Normal. It was a
completely wrong answer, but so clever and humorous that the student
survived.)
Reviewing the calculations for the upper and lower control limits for both the
mean and the range in Table 1, it will be noted that the products of the factor
values and the calculated sample range can be used to determine these limits.
Sample number
1 2 3 4 5 6 7 8 9 10 11 12
1 3.50 3.60 3.60 3.60 3.50 3.70 3.70 4.00 4.50 3.50 3.70 3.70
2 3.50 3.50 3.70 3.70 3.50 3.80 4.00 4.40 4.60 3.60 3.70 3.60
3 3.60 3.50 3.50 3.60 3.60 3.90 4.20 4.50 4.20 3.60 3.60 3.80
4 3.50 3.60 3.60 3.60 3.60 3.90 4.10 4.20 4.10 3.60 3.60 3.80
Total 14.1 14.3 14.4 14.5 14.2 15.3 16.0 17.1 17.4 14.2 14.6 14.9
x 3.525 3.55 3.60 3.625 3.55 3.55 4.0 4.275 4.35 4.55 4.65 4.725
High value 3.60 3.60 3.60 3.60 3.60 3.60 3.60 3.60 3.60 3.60 3.60 3.50
Low value 3.50 3.50 3.50 3.50 3.50 3.50 3.50 3.50 3.50 3.50 3.50 3.50
Range 0.10 0.10 0.20 0.10 0.10 0.10 0.50 0.50 0.50 0.10 0.10 0.20
Table 1. XyR data and for the dimension A

calculations
X= 2XIn= 45.225/12 = 3.769

R= Z RJn = 2.7/12 = 0.225
LCSX = X + A 2 R = 3.769 + 0.73(0.225) = 3.93
LClX = X- AR= 3.769 - 0.73(0.225) = 3.60
LCSR = D 4 R = 2.28(0.225) = 0.51
LClR = D 3 R = 0(0.225) = 0

2 7
8
The values for these factors, for various sample sizes, are given in Table 2.
Sample size TO D3 D4
2 1.88 0 3.27
3 1.02 0 2.57
4 0.73 0 2.28
5 0.58 0 2.11
6 0.48 0 2.00
Table 2. Factors for determining upper and lower control limits for XyRa plots from R
When analyzing the plotted means for the diameter A values for the first 48 cut
pieces (Fig. 2) It is evident that this process appears to be "out of control,"
meaning that too many (i.e., more than 1 in 100) sample mean values fall
outside the control limits. This suggests that there is a special underlying cause
for the displayed degree of variability of the sample mean. The suspicion here is
that something is wrong with the way this operation is performed.
The underlying ability of the cutting operation to maintain the dimensions of the
diameter value is much higher than that reported in the sample mean values.
Numerous questions about the process should be considered. For example,
does the carousel in the volume hold the piece firmly? Is the tool blunt and
therefore causing an uneven cut that unbalances the diameter for some portion
of the tube? Is there material inconsistency along the tube?
Department Lathe 3058 Machine Part Name axis

No. part cut diameter Variable Browne & Sharpe
Operation diameter 3.932775
Grand average 3.768750 LCl x Bar 3.604725 LCS x Bar
Medium range 0.225000 LCS Range 0.513000
x Bar graph
0.00
Fig. 2 Measurements plotted for diameter values A

2 7
9
Consider another example, shown in Table 3, regarding the pitch diameter of

threads in aircraft fittings.
Table 3 Diameter of thread pitches in aircraft fittings
Measurements of the pitch diameter of threads in aircraft fittings. Values are expressed in units of 0.0001
inches in excess of 0.4000 inches. The dimension is specified as 0.4037 ± 0.0013 inches
Measurement on each item of five items per Average X
Sample number R Rank
hour
1 36 35 34 33 32 34.0 4
2 31 31 34 32 30 31.6 4
3 30 30 32 30 32 30.8 2
4 32 33 33 32 35 33.0 3
5 32 34 37 37 35 35.0 5
6 32 32 31 33 33 32.2 2
7 33 33 36 32 31 33.0 5
8 23 33 36 35 36 32.6 13
9 43 36 35 24 31 33.8 19
10 36 35 36 41 41 37.8 6
11 34 38 35 34 38 35.8 4
12 36 38 39 39 40 38.4 4
13 36 40 35 26 33 34.0 14
14 36 35 37 34 33 35.0 4
15 30 37 33 34 35 33.8 7
16 28 31 33 33 33 31.6 5
17 33 30 34 33 35 33.0 5
18 27 28 29 27 30 28.2 3
19 35 36 29 27 32 31.8 9
20 33 35 35 39 36 35.6 6
Totals
... ... ... ... ... 671.0 124
Figure 3 illustrates values for individual members of sampled diameters in

relation to the nominal dimension of the component and the upper and lower
tolerance limits. Of the 20 samples of 5 readings each (i.e., 100 individual
diameters measured), only one diameter on sample 8 (i.e., 0.4023 inches), falls
outside the tolerance limits. Is the process in control? That depends on what
control means.
Figure 4 illustrates X and R control charts for the 20 samples. The control chart
limits, upper and lower, are determined statistically such that there should only
be a chance in 100 that a sample mean will fall outside these limits. Note that
three sample means are out of bounds for .
In other words, something is probably happening that shouldn't. In quality

control language, the process is not in control. Recall that 99 of the 100 parts
measured fall within tolerance limits (i.e., are acceptable).

3 7
0
Sample number
Fig. 3 Individual diameter measurements
0.0020
x
L' half of control upper

0
— ——
5 -X----- x
8
5
either
3 0.0010 x
Q.
K >
0 -------------
5
c
6
K
XX Yo
XX x
XX
Limit
X for
cont
inferior x
role or
—
0
5 10 15 20
Sample number
Fig. 4 X and R control charts for feed diameter

3
1
Simply put, the range of part dimension acceptability, in terms of use, is

so wide that even when the process is "out of control" it is still capable of
producing acceptable parts (99 out of 100 times). With better process
control (remember: there is reason to believe that something may be
producing undue variability) this process would have a greater ability to
successfully produce the part.
A common adjustment for a process based on information from such

charts is to alter the frequency and degree of centering or positioning of
the process (for example, remounting it as close to 0.4037 inch as
possible, as often as necessary, if it is deviating from the center position;
or move it down within the acceptable range as the tool wears).
The reason for plotting sample means rather than individual values goes
back to the desirable properties of the central limit theorem, which will be
examined. Suffice it to say that the reliable determination of control limits
depends on the assumption that sample means are normally distributed,
independently of the distribution of the individual values from which the
samples were obtained.
X and R charts have been used in statistical quality control since
Shewhart [14] introduced these concepts in the early 1920s. Over half a
century and into the 1980s, statistical quality control developed to include
a considerable diversity of statistical techniques for managing quality
problems in manufactured products. Finding the source of quality
problems was one part of the story, but it was generally assumed that
such problems in the manufacturing process could be resolved once they
were discovered.
The use of X and R charts and other related statistical approaches to find
the causes of quality problems on the shop floor was, unfortunately, a
case of "too little, too late."
An important point must be highlighted. Far more important than the

most complex mathematical formulations we can conceive, or the
calculations we can perform (feats with which students have become
intimately associated during their engineering studies), are the guiding
principles, concepts, or premises found in the core of the practice of a
profession. Quality control as practiced from 1920 to 1980 had a very
high statistical basis, if not in practice, at least in theory.
In retrospect, it is clear today that although much was achieved, much

more could have been achieved with a better understanding of the true
underlying source of quality problems in the industry. A review of the
deduced definitions of quality provides some insight into this gradual shift
in approach to more successfully meet the challenge of improving quality
in the future.
TOTAL QUALITY
Basic philosophy of quality
1.1 Introduction
We are practically in the middle of the last five years of the 20th century
and the current times are certainly increasingly difficult for every
3
2
organization, as we approach the third millennium, due to the process of
accelerated change and global competitiveness that the world is
experiencing, where the liberalization of economies and free competition
come to characterize the environment of inexorable coexistence for the
business sector.
In this context, companies have to continue assuming the role that

corresponds to them to contribute to the growth and development of the
country, achieving greater efficiency and providing quality products and
services. Today, more than ever, there seems to be a very broad
consensus regarding the urgent need for companies to function well
competitively.
Until a few years ago, the protectionist system in our country, as in other
Latin American countries, had prevented us from assessing the harsh
conditions of international competition and the higher levels of demand
from clients and consumers, who demand greater quality in products,
opportunity in deliveries, reasonable prices and excellence in service.
The harsh reality that began in the 1980s and the effects of globalization
in the 1990s is abruptly awakening all organizations and forcing them to
eagerly seek new strategies to successfully adapt to growing
competition.
It is precisely in this environment in which Total Quality is vigorously and

revolutionaryly projected as a new business management system and a
first-order factor for the competitiveness of companies. The concept of
quality, traditionally related to product quality, is now identified as
applicable to all business activity and all types of organizations.
Successful companies in Peru and the world are those that have been
applying the Total Quality strategy in one way or another; and many
others based on this strategy are changing their way of thinking and
therefore acting; In fact, they are redesigning the entire organization,
both physically and spiritually, to focus it on customers, and make it
efficient to fulfill and satisfy them.
Many of our companies, although they recognize the importance of

quality, are not sufficiently prepared to accept the new challenges that it
brings with it and to put its principles and techniques into practice.
Perhaps one of the main drawbacks is the lack of a practical
methodology to support them.
In the session corresponding to this first day of work we will present a
review of the general concepts and principles on which the philosophy of
Total Quality is based.
We hope that the initial discourse, complemented by the theoretical-

practical instruments of the following days, will guide the reader's way of
thinking, and that to the extent that this discourse penetrates their mind,
the application to their concrete reality will become possible, or will
contribute. to direct total quality processes already started.

3 7
3
1.2 Concept and evolution of quality
Concept:
An essential condition to ensure the implementation of a Total Quality

strategy is to clearly define and understand what this concept means.
That is, the managers of an organization that intend to implement Total

Quality as a strategy to compete have to know exactly what they mean
when they talk about quality, or about improving the quality of the
product or service, they have to know how to divide global quality. of
manageable improvement projects and how to measure product quality.
In practice, as Richard J. Schonberger, one of the experts in this

matter,"...quality is like art. "Everyone praises her, everyone recognizes
her when they see her, but everyone has their own definition of what she
is."
Dictionaries provide us with a first basis for the formation of this

theoretical framework. One of the definitions that we find in the Little
Illustrated Larouse tells us that Quality is the quality of a thing. Another
meaning collected from the dictionary is that quality means a way of
being of a person or thing.
In these definitions, quality is implicitly conceived as an attribute,

property or characteristic that distinguishes people, goods and services,
which is already an interesting approach to the concept of quality applied
to organizations.
The international standardization body, ISO, in its standard 8402, has
defined quality as the totality of characteristics of an entity that gives it
the ability to satisfy explicit and implicit needs.
The standard specifies which entity is an organization, whether it is

called a company or institution, product or process. Complementing this
definition, we will say that explicit needs are defined through a
contractual relationship between Clients and Suppliers; while implicit
needs are defined according to the conditions that prevail in the market.
The elements that make up the needs are basically: security, availability,
maintainability, reliability, ease of use, economy (price) and the
environment. These needs, except price, are defined by translating
aspects and characteristics necessary for the manufacture of a good
product.
Quality Evolution
In general, it can be said that the concept of quality and its application,
until reaching the current state, has had the following evolution:
a) Quality control focused on finished products.

b) Statistical processes control.
c) Total Quality Control or Total Quality.
The first stage that began with the industrial revolution consisted of the
inspection of finished products, classifying them as approved or rejected.

3 7
4
The latter had to be subjected to reprocessing if possible or simply
eliminated.
In this traditional conception, quality is normally associated with a
production chain and is often seen as competing with other business
priorities such as cost and productivity reduction. To increase
productivity, quality had to be sacrificed.
Another problem with this traditional concept of quality is its focus on

correcting errors after the fact; This check-and-fix-later philosophy not
only allows for errors but also incorporates them into the system. Our
experience as Customers is that too many times we end up buying the
errors that result from checking and fixing later. Many times there are
more possibilities of purchasing defective goods and services than
perfect ones. But if that happens to us as Clients, the question is, what
will ours say?
The final problem with this approach is that it is very expensive to fix
things that have gone wrong. Well, the more you try to improve with
traditional quality, the more expensive it becomes. Traditional quality is
something imprecise, we all think about it in different ways, it takes a
secondary place before other objectives such as productivity. It not only
allows for errors but incorporates them into the system through the
philosophy of checking and fixing later. This way of doing things costs
companies a lot of money and causes them to lose Clients.
The second stage, begun in the first half of this century, consisted of the
development and application of statistical techniques to reduce
inspection costs. With this approach, it was possible to extend the
concept of quality to the entire production process, achieving significant
improvements in terms of quality, cost reduction and productivity. The
advantages offered by Statistical Control allowed it to expand its
application to other areas of the organization; However, it was noted that
although this method tremendously improved the company's results, it
was insufficient to face the growing competitiveness.
This is how Total Quality Control and the idea of Continuous

Improvement were born, as a way to be successful on the journey
towards excellence, that is, to achieve Total Quality.
This concept was born in the 1950s in the United States, but it was in
Japan where it was fully developed and applied, introducing important
and novel concepts such as:
• Quality means satisfaction of customer needs and expectations.

• The conception of internal clients and external clients.
• The responsibility of management in quality.
• Quality should not only be sought in the product but in all functions
of the organization.
• The participation of staff in permanent quality improvement.
• The application of principles and tools for the continuous
improvement of products and services.

3 7
5
1.3 Understanding the meaning of total quality
The concept of Total Quality, originated from the expanded concept of

quality control (Total Quality Control), and which Japan has made one of
the pillars of its industrial renaissance, has made it possible to
standardize the concept of quality by defining it according to of the client
and thus avoiding diversity of points of view as happened in the
traditional conception. This is how Quality becomes total.
Quality is total because it includes each and every aspect of the

organization, because it involves and commits each and every person in
the organization. Traditional quality was about fixing quality after making
mistakes. But Total Quality focuses on getting things done right the first
time. Quality is built into the system. It is not an afterthought and the so-
called acceptable quality levels become more unacceptable every day.
Complementing what has already been mentioned, we must say that

Total Quality is meeting the requirements agreed upon with the client
and exceeding them, now and in the future, we must start by being exact
with the requirements or specifications. With this concept of Total
Quality, the imprecision of the past is overcome; it not only tends to be
exact but also measurable. Another change that is obtained as a result of
this conception is the word Client.
Customers are no longer just the ultimate Users of the goods and
services we sell, now the term is expanded to include the idea of Internal
Customer, the people in the organization to whom we pass our work.
With this concept obviously everyone in the organization becomes
someone's client; What's more, it acquires a dual character of being
Client and Supplier at the same time.
Perhaps to better understand the concept of Total Quality, it is

convenient to start by saying that the objective of every organization,
work group, area or job, or even the individual, is to generate a product
or service that another organization, another area or other area will
receive. another individual, whom we call, also called User or Consumer.
Some prefer to call Customer when it is a tangible product such as a car

and User when it is a service, such as health or repair service. In our
case we will call Customer, whether it is a product or service.
It is worth specifying that the term product refers to the result obtained
from a process or an activity. Therefore, in general terms, this result can
be a tangible product (for example, assembled or processed materials),
or intangible (for example, knowledge or concepts) or a combination of
these; Product is the work performed by a job or workstation. It is also a
product of the result of advice, or an educational service, etc., in general
of all those activities where the result obtained is not perceived as a
tangible product.
However, for the purposes of Total Quality, the term service has been
defined as the result generated by activities at the interface between the
supplier and the client and by internal activities of the supplier, in order to
know the needs of the client.

3 7
6
Let's see for example, in the case of the Toyota Company, customer
satisfaction is classified into two aspects: product and service. When
they refer to the product they mean: quality, reasonable cost and
timeliness of delivery. On the other hand, service for this company
means communication and permanent contact with the client, as well as
attention, which can even be developed perhaps to recreational activities
intended for clients; Likewise, part of the service is keeping all
merchandise in contact with the customer, post-sales monitoring to verify
the degree of satisfaction with the product, and the entire set of
interrelationships or activities to capture the customer's voice and satisfy
them better.
Other companies include as part of the service the so-called "pre-sales

service" aimed at making the benefits of the product and the company's
image known to people who are not customers, who perhaps could be,
but not necessarily.
From the point of view of Total Quality and accepting the ideas
presented in the previous paragraph, the concepts of product and
service are not separated, either the product includes the service, or both
aspects (product and service) must be planned separately and
simultaneously to achieve better satisfaction of customer needs and
expectations.
In a simple way we can say that in the expression Total Quality, the term
Quality means that the Product or Service must be at the level of
satisfaction. the client's; and the term Total that said quality is achieved
with the participation of all members of the organization. Total Quality
involves a series of innovations in the area of business management that
has spread throughout all the countries of Europe and America, applying
not only to manufacturing activities but also to the service sector and
public administration.
Today, in Peru, there are many organizations that have been

successfully applying this new strategy.
Total Quality means a paradigm shift in the way an organization is
conceived and managed. One of these fundamental paradigms that
constitutes its reason for being is constant improvement or continuous
improvement. Total Quality begins with understanding the customer's
needs and expectations and then satisfying and exceeding them.
1.4 Strategic importance of total quality
Total Quality is a strategy that seeks to guarantee, in the long term, the
survival, growth and profitability of an organization by optimizing its
competitiveness, through: permanent assurance of customer satisfaction
and the elimination of all types of waste. This is achieved with the active
participation of all staff, under new leadership styles; Being the strategy
that, well applied, responds to the need to transform the products,
services, processes, structures and culture of companies, to ensure their
future.
To be competitive in the long term and achieve survival, a company will

need to prepare with a global approach, that is, in international markets

3 7
7
and not only in regional or national markets. Well, being excellent at the
local level is no longer enough; To survive in today's competitive world it
is necessary to be successful on the world stage.
To successfully adopt this strategy, it is necessary for the organization to

implement a permanent improvement process. The essential aspects for
the application of this process will be presented in the second session.
1.5 Total quality as a new business management system
Total Quality constitutes a new business management system, to the

extent that its concepts radically modify the characteristic elements of the
system traditionally used in Western countries. Among them are:
■ The values and priorities that guide the management of the

company.
■ The logical approaches that prevail in the management of business
activity.
■ The characteristics of the main management and decision
processes.
■ The techniques and methodologies applied.
■ The climate understood as the set of perceptions that people have
about relationships, personnel policy, environment, etc.
The new quality approach requires a total renewal of people's mentality

and therefore a new business culture; since among other aspects,
participatory management and a revaluation of personnel not applied in
traditional modes of administration must be put into practice.
1.6 Reflections for the reader

In order to help the reader identify opportunities for improvement, we
present below a set of reflections that we consider important, the same
ones that we have adapted from Jaime Rojas' book referring to the
application of Total Quality.
■ Is your organization prepared to satisfy customers against the best
competitors?
■ Do the products and services you offer respond to the true needs and
expectations of customers?
■ Are you eliminating all types of waste to have competitive costs and
adequate profitability?
■ Is the timing of your development cycles for developing new products
or services and those for operation, delivery and after-sales
adequate?
■ Is your staff highly motivated and actively participating in process
improvement?
■ Have you managed to develop reliable suppliers?
■ How far are you currently from Total Quality?

3 7
8
The improvement process towards total quality
2.1 Basic principles for achieving total quality
1. Quality is the key to achieving competitiveness.

With good quality it is possible to capture a market and stay in it.
2. Quality is determined by the customer.
It is the client who qualifies the quality of the product or service

offered; Hence, quality should not be taken in its absolute or scientific
value, but rather is a relative value, depending on the client.
It is necessary to accurately identify the changing needs and
expectations of customers and their degree of satisfaction with the
company's products and services and those of the competition.
Keep in mind that customer expectations are given in terms of quality

of the product or service, timeliness of delivery, quality of care,
reasonable costs and safety. Don't ask yourself about the quality of
the product or service you offer, ask your client, he is the one who
tells what to do, how or when.
The Customer cannot be forced to buy the product we want at the

price we want. An organization improves toward Total Quality when
external and internal customers feel that their quality, timeliness, cost,
and service requirements are being consistently met.
3. The production process is throughout the organization.
Production process is not the entire production line itself, but the
entire company.
Those who do their job well lubricate the process, those who do it
poorly create bottlenecks in the process.
4. The quality of products and services is the result of the quality of

the processes
5. The supplier is part of our process.

Quality begins with the supplier, he is part of our process, he must be
considered as part of the organization.
Quality begins with demand (from our customers) and will culminate
with their satisfaction, but the production process begins with the
supplier; Therefore, this must be considered as part of our production
process, extending to it the Total Quality training actions.
6. Internal supplier-customer chains are essential.
Each individual in the organization becomes aware that they have one
or more internal customers and one or more internal suppliers.
Creating supplier-customer chains within the organization. Internal
suppliers who must be kept informed of how we want their work to be
delivered to us and what needs to be corrected.
In this chain that is generated by this awareness of internal suppliers

3 7
9
and internal clients, the supplier who does not ask his client about the
quality of the work he delivers is as responsible as the client who does
not inform his supplier about something that It is being delivered badly
done.
7. Quality is achieved by people and for people.
This requires, among other things, a constant training and training

program.
8. Establish the zero defect mentality.
This is intended to eradicate waste, in all its forms, by eliminating

activities that do not add value.
"Zero defects" consists of having a systematic attitude towards no
mistake. We owe the common use of the phrase to err is human. It is
not about persecuting subordinates because they make mistakes,
since that would kill their initiative.
What it is about is awakening the awareness of not making mistakes.

Under the concept of ZERO DEFECTS, tolerance limits will disappear,
as they enshrine error.
Total Quality promotes the elimination of all types of waste present in:
- Inventories.
- Equipment not available due to damage or maintenance.
- Personnel dedicated to repetitive or inofficious tasks.
- Papers and excess procedures.
- Excessive reports and meetings.
- Work inventories in interoffice processes.
- Unnecessary internal controls.
9. The competitive advantage is in the reduction of errors and

continuous improvement.
The true competitive advantage is in reducing errors or improving
processes; Therein lies the cost reduction. With the result of "Lower
Costs" you can:
- Lower prices for customers.

- Improve company profits.
- Improve the finish of the product.
- Or all of the above at the same time.
10. The participation of everyone is essential (collective

consciousness).
Reducing errors will only be possible with the participation of each
and every member of the organization.
ll.Quality
It is above all a managerial responsibility. Managers must be leaders.
capable of involving and committing staff in improvement actions.
12. It requires a new culture.

In the concept of Total Quality "everyone thinks and everyone does"
4 7
0
2.2 Improvement Modalities
Within the Improvement of a company, two main modalities are
distinguished: Continuous improvement and innovation or Radical
Improvement.
a) Continuous improvement
They are incremental contributions that are achieved with the
participation of all staff, motivated by a challenge of permanent
improvement, known by the Japanese name of Kaizen.
b) Radical innovation or improvement

It refers to radical contributions or large changes made sporadically
by specialists in the respective topics. An example of this modality
is called "Process Reengineering". Radical improvement is the
responsibility of Senior Management and is achieved with important
changes and technological innovations that involve large
investments, while incremental improvement, Kaizen, is a process
of continuous renewal. carried out through small steps that serve to
perfect existing standards, being in charge of all levels of the
company.
The ideal improvement is one that effectively combines continuous

improvement and innovation; it is the best way to ensure the future
success of a company. However, we must point out that the application
of Total Quality means a radical change in people's mentality, and the
holistic conception of the organization and improvement processes that
characterizes this approach plays a fundamental role for innovative
purposes; Well, not only does it promote innovative capacity, but it also
discovers where it is necessary to innovate, because the optimization
work has reached its limit.
In this way, Total quality is decisive for innovation that also becomes
global, taking root in the company with the additional ingredient of
participation.
2.3 The control cycle for improvement
Quality improvement and assurance is based on the Control Cycle,

which is made up of the following four phases:
■ PLAN: Activity that determines what should be done, what the goal
is and how it can be achieved.
■ DO: It means transforming current processes in order to improve
their performance, according to plan.
■ VERIFY: Determines the degree of compliance with planned
activities and performance goals.
■ DO: It means making adjustments to new procedures and
standardizing them, in order to ensure that they are always
applied.
The continued execution of the Control Cycle guarantees the permanent

improvement of the performance of the organization's processes.

4 7
1
2.4 Activities to start a process towards total quality
For the implementation of Total Quality there is no model that can be

copied. In principle, each organization has to design a methodology
according to its own reality.
As part of any improvement process towards Total Quality, actions

are included such as: the development of leadership activities and
the commitment of the highest managers of the organization, the
development of personnel, the focus on customers, the development
of suppliers, quality planning, daily work improvement, quality
assurance, cost reduction program, etc.
Based on the recommendations of experts in this matter, our experience

and our own empirical evidence, we would like to point out that in
general terms, to put this process into practice it is convenient to divide it
into the following phases:
PHASE I: Decision making
It consists of seriously analyzing whether there is the will to do
everything necessary to adequately undertake and maintain said
process; since once started it should not be discontinued as it generates
mistrust and frustration.
It is necessary to be aware of the efforts it will take to overcome
resistance to change. On the other hand, managers have to dedicate
time and resources, guide administrative management, accept all
concepts and principles such as: teamwork, new leadership styles, have
a clear definition of quality, etc.
It is necessary to be aware of what it means to undertake a process
towards Total Quality and the benefits that derive from it.
Once the decision has been made, a Plan must be developed to launch
this process. This Plan must be integrated or form part of the
organization's Strategic Plan. Likewise, once the decision has been
made, a basic organization that will be responsible for developing quality
actions must be defined in parallel with the Quality development plan. As
part of this organization, a Quality Committee or rather a Quality Council
will be included, chaired by the highest manager of the organization. then
the Quality Committees by Area and Improvement Teams depending on
the size of the organization. The need to hire the services of an External
Advisor should also be taken into account. The quality plan must define a
pilot area to initiate improvement processes and also provide for
personnel recognition actions.
PHASE II: Stage preparation and promotion
To begin this stage it is assumed that there is already a Plan in which all
the actions to be developed have been disaggregated. Some of the
actions that are recommended to be carried out at this stage are:
develop an intense program to disseminate the concepts and philosophy
of quality to internalize it in people as a means of achieving their
involvement and commitment. The vision must be developed. mission,
policies and objectives of the organization, which will be disseminated
among all staff during quality dissemination and promotion actions.

4 7
2
It is convenient in this phase to carry out a Quality Diagnosis of the
organization, considering aspects such as costs of poor quality,
organizational climate, level of customer satisfaction, identification of
critical processes, clarity and precision regarding strengths and
weaknesses.
In this phase it is necessary to launch a vigorous Training and Personnel
Development Program. The development of projects will be tested at a
pilot level.
PHASE III: Implementation of improvement processes
In this phase, the improvement processes will be launched with the

participation of the improvement teams, who by this time have been
sufficiently trained in improvement and problem-solving techniques.
If the improvement process is showing its fruits, it is necessary to

motivate staff with recognition actions. The training aspect must include
the application of ISO 9000 series standards.
PHASE IV: Consolidation and interfunctional optimization
In this phase, improvement processes will continue to be developed and

training in more advanced quality techniques will continue. Systems and
procedures will be developed based on ISO 9000 standards. The
improvement processes will continue, maintaining and perfecting
strategic management.
The above is not definitive or absolute; it will depend on the dynamism

and progress that each organization has achieved. Without entering into
controversy with the above, an operational methodology that has been
followed by an organization in our country where we have had the
opportunity to participate and provide advice consists of the following
actions:
1. Decision making: Situational analysis; analysis of quality approaches

in the world; analysis of the experiences and achievements obtained
by other organizations.
2. Design, structuring and approval of the total quality management

plan.
3. Implementation, at the level of a pilot area.
4. Expansion to the entire organization.
5. Evaluation and monitoring.
6. Corrective actions, assurance measures and continuation of

strategic management.
7. Development of assurance and audit actions for accreditation

purposes.

4 7
3
QA
The concept of Quality Control comprises two interrelated functions:
a) Control function
b) Quality function
For our purposes, the control function includes at least the following
elements and activities:
- Set the standard

- Measure the phenomenon
- Compare the measurement with the norm
-Analyze the importance and characteristics of the differences.
- Establish the foundation for corrective action.
For its part, the quality function has at least the following elements or
stages:
-Design quality
- Production quality
- Compliance quality
- Certified quality
-Quality of service
The interrelation of these two functions constitutes in its broadest form

the modern concept of quality control as seen in Fig. 1.
QA
Fig. 1
1. Functions and principles of quality control
FUNCTIONS: Three functions are distinguished:

4 7
4
* Legislative
* Executive
* Judicial
The Legislative Function involves the creation of control standards,

preparation of specifications, etc. In a word, he makes the law. It is
characteristically the stipulation of design quality. This is not an exclusive
function of Quality Control, as other departments are involved, but a
design should never be carried out without the intervention of the Quality
Control Department.
Within the Executive Function are all those tasks that make the process
deliver products according to the design, avoid the manufacture of
defective parts and enable the location of waste sources, that is, what we
will call control during the process. of manufacturing.
In the Judicial Function we will find entry and exit control, that is, the
judging of already finished products, which can be raw material for other
sectors or factories.
So, summarizing, we have that design quality is in the legislative function
and product quality in the executive function, dynamic and judicial quality
control, static quality control.
Principles (Refers to product quality control)
1. "Control is not classification or selection"
In effect, the classification of products into good and bad is just

another manufacturing operation. Quality Control must provide
reasons for studying quality problems.
2. “With control you cannot obtain quality; This is an inherent

characteristic of the product."
This is evident; To obtain a good level of quality it must be

manufactured. Control does not add quality to products.
3. "The production team is responsible for quality and control"
This principle is a consequence of the previous one, and from it it is

concluded that we must provide the production team with the powerful
tool that is control during the process, which guides them to know the
corrective measures they should take.
4. "Control does not solve manufacturing problems, it only gives

reasons to study them"
It is very important that the production team knows what quality

problems exist and in what sense they manifest themselves to
achieve a good level of quality in production.

4 7
5
5. Decisions must be made based on real data."
The reliability of the recorded data is the starting point for any analysis
or interpretation of results. This task is typical of quality control and
the production team will take corrective measures based on this
objective data.
6. "The data must be compatible and arranged in such a way that it

allows analysis."
This will make it possible to use some statistical tools that quality
control makes use of.
7. "The agreement between suppliers and customer must be made

on the control method and only on the defective fraction of the
losses."
There is what is called buyer's risk, which, in general terms, can be

defined as the risk of accepting items with a defective percentage
greater than that specified and the supplier's risk that items with a
defective percentage lower than that specified are rejected.
Always keeping these basic principles in mind and ensuring that all
parties involved understand them, it is certain that the efforts we will
make to achieve the desired quality levels will not be distorted.

4 7
6
QA
Why produce quality products? (Fig. 2)
The company must produce quality products by:
- Financial reasons; Quality defects mean a very high cost for both the
company and the customer.
QUALITY DEFECTS
THEY COST A LOT
To the
enterprise
To the
client
QUALITY DEFECTS =
WASTE Work times and
unavailability
Raw materials and

manufactured products
Energy
Sales price increase

COMPANY Decrease in profits
Loss of
competitiveness

4 7
7
- Commercial Reasons; need to be competitive Customer
(Competitiveness = Quality/Price)
At the same sales price
Better utilization
Better availability
MOST
SATISFIED
CUSTOMERS
Better adaptation to your needs

Increase quality
- External demands; made up of associations of: consumers,
environmental standardization and public authorities.
They are due to:
a) Main industrial clients

Those who request certification of the demanded quality.
b) Consumer society
Those who report, through comparative tables, the characteristics,
performance and price of the products. Reception of complaints and
consumer defense.
c) Public powers
Those who publish the regulations referring to the safety of use.
d) Standardization institutes
Solely responsible for product quality compliance.
e) Environmental associations
Responsible for avoiding environmental pollution.
- Internal Requirements
Improvement of the company's work environment.
The demotivation of workers is reflected with the following behavior:
- A high level of absenteeism and tardiness.

- A high turnover rate.
- High accident rate.
- Poor quality and high rate of waste in products.
- Low productivity.
- An inflexible workforce.
- Low morale and general dissatisfaction.
"TO REMEDY IT, THE SPIRIT OF QUALITY MUST BE DEVELOPED"

4
8
EQUAL QUALITY
Reduce the costs of:
NEED rejection, repairs,
maintenance and controls
OF
Streamline work methods
DECREASE THE PRICE
and manufacturing
processes
SELLING
Simplify product design
Build customer loyalty

MAINTAIN OR IMPROVE THE BRAND IMAGE TO
Expand markets
- Technical reasons; improve properties and performance linked to use;

as well as mastery of manufacturing and control procedures.
To avoid the negative consequences of financial and business

reasons, we must improve:
For more economical

THE use and better adapted
PERFORMANCES
TECHNICIANS
to use
Reliability
THE PROPERTIES Maintenance
LINKED TO THE
UTILIZATION Lifespan
THE DOMAIN OF Manufacturing processes and

TECHNIQUES
CONCERNING control methods
STANDARDIZATION AND QUALITY SPECIFICATIONS

4 7
9
Introduction to standardization
In today's society, commercial relationships are growing rapidly, where
comparators and sellers require better processes that allow these
transactions to be carried out with a high degree of reliability with
respect to the quality of the products that are sold. Quality is not an
absolute magnitude, it is influenced by variable subjective conditions,
which needs to have a system that serves as a connection between
buyers and sellers, to generate technical bases for good commercial
understanding, such as metrology, standardization and Quality Control.
It is important to specify the functions that the good or service to be
purchased will fulfill and thus know the potential that the manufacturer
has to respond to customer expectations. This generates the
appearance of different levels of quality and prices. The success of all
these aspects to ensure the expected qualities lies in the fact of a set of
technical rules, recognized and followed that make up a system of
standards. This system of understanding and control is called
normalization.
The system that makes up standardization is made up of a
conceptualization, differentiation and ordering of a series of principles
and procedures to establish units and definitions, methods of taking and
preserving samples, test methods, as well as qualitative specifications
and codes of practice. .
For Engineer Rafael Salas Jiménez, from the technical department of A
VPC, "The problem is that to achieve sufficient roots and extension of
standardization, it is necessary for society to be truly aware of this
need." This suggests that the first task is to create awareness of the lack
of a standardization and quality control process, where it is necessary to
demonstrate that these changes are necessary, so that the modification
of behavior is not temporary.
Standardization aimed at manufacturers is called industrial
standardization, where it is an important particular, restricted and
restrictive case that is responsible for bringing to the organization the
foundations for a harmonious and efficient development of the industry.
The degree of evolution of standardization in a specific area determines
its degree of importance and industrial development.
In the vast set of buyers' markets, the industrial buyer has advantages
over the common buyer, since suppliers are subject to the industrial
standardization that the organization exercises in its purchases, based
on guarantee agreements, specifications and agreements that are not
present. in the common buyer. The average buyer may not know exactly
which products best meet their particular needs. For this reason,
standardization has the important task of developing the production
items that go on the market, and helping to define a type of quality for
those buyers who lack the necessary technical knowledge.

5 7
0
It is likely that there is no complete feedback between complete, correct

and sufficient information about a specific product and its sellers or
manufacturers. Standardization tries to avoid these peculiarities by
defining the procedures and standards to be used in the work tables and
in their preparation in the long and difficult technical discussions.
A poor relationship between buyer and manufacturer and the absence of

mechanisms that guarantee the reliability and quality of products and
materials, leads to companies carrying out quality tests on their own on
materials from the same supplier, and where buyers duplicate the tests
and controls on the same products from a given manufacturer.
These controls are the following:
Production or internal controls carried out by the manufacturing or

service company.
Acceptance or rejection controls, external, carried out by the buyer.
1. Standardization
General concepts
The need to regulate various human activities dates back to the

beginnings of civilization, due to the tendency of problems to repeat
themselves; It is then that solutions are proposed to these repetitive
problems, these solutions tend to become norms, written or not;
Regulations, rules or norms arise, for example, how to communicate,
how to take care of oneself, how to cultivate the land, etc. The evolution
of these rules, particularly those of communication, lead to language and
writing, grammatical rules, ways of conjugating verbs, the way of forming
plurals, the use of "S", "C" or "Z", etc. . They are normalizations.
Commercial application standards, standards for measuring lengths,

areas, volumes, weights, etc. are also developed.
When a specification produces satisfactory results, we will use it every

time we need the same result. This repetitive use of a specification is a
STANDARD.
So; We can mention some facts that mark important steps in the field of
industrial standardization.
It makes all the parts of a weapon 1793 AWWHITUEY interchangeable,

that is, the components of one weapon make it fit another of the same
type.
Establishes standards for screw threads, 1841

WHITWORTH bolts and nuts that are held up to
present.

5 7
1
1846 The separation between railway rails is unified.
In our century, normalization takes the great leap from empiricism on the
one hand and regionalism on the other; that of precision, technology and
universalization.
Organizations appear, at national and international level, that are

dedicated almost exclusively to standardization.
BSI BRISTIHSTANDARD INSTITUTION 1901

SA International Federation of National Standardization Associations
1926-1939
(It works until the start of the 2nd. World War)
ISO International Organization for Standardization 1947
In our country:
1959 INANTIC - National Institute of Industrial Technical Standards and
Certification is created.
1970 The lTINTEC Institute of Industrial Technological RESEARCH and

Technical Standards is created.
1992 The National Institute for the Defense of Competition and the
Protection of Intellectual Property is created.
The functions of these entities, which have occurred over time, are
generically and among others, to promote, study, establish, review,
verify and certify technical standards; support all sectors of the National
Economy in aspects related to normalization.
STANDARDIZATION
Definition according to ISO
"Normalization is the process of formulating and applying rules with the

purpose of bringing order to a given field of activities, for the benefit and
with the participation of all those involved and, in particular, to obtain an
optimal economy of sets, respecting the functional and safety
requirements.
It is based on the consolidated results of science, technique and

experience; “It not only sets the foundations for the present but also
projects them for future development, with which it must march in
unison.”

5 7
2
It is a discipline that deals with the establishment, application and
adaptation of a set of rules intended to achieve and maintain an order
within a certain field, in order to provide an Institution, Company, the
country or the International Community, certain economic benefits. or
social.
It must have a defining characteristic: it is dynamic; that is, in a constant

process of updating.
Technical rules
"Result of a particular or specific standardization work"
It is a pattern or model, a set of reference data, it is concrete and

precise, it is materialized in a written document where it is fundamentally
specified:
- The name of the object, good or service, clearly differentiated and

individualized from any other.
- Requirements and/or numerical values, which it must have to adapt

to its use functions.
- Test methods to test or verify these requirements. These tests may

request separate standards.
- Identification, labeling, packaging and/or packaging, if applicable.
A Technical Standard can also manifest or be specified as a device or

equipment, an instrument, or an object such as a viscometer, a
thermometer or a ruler - meter, respectively.
Types of technical standards
By the nature of its application:
• Mandatory standards
• Optional or voluntary standards
For its content:
• Terminology standards and definitions

• Classification rules
• Preparation rules
• Requirements standards
• Test method standards
• Sampling and reception standards
• Labeling standards
• Packaging and packaging standards
• Rules of use.
When a specification gives satisfactory results, we will use it
1. Every time we need the same result. This repetitive use of a

specification is a STANDARD.

5
3
2. Conditions that a Standard must meet:
a) That responds to a real need.
b) That it be prepared with the participation and consensus of the

interested groups.
c) Make it quick and easy to apply.
d) That promotes or causes economy, cost reduction and/or

improvement of the product or service.
e) That it be reviewable and perfectable when the conditions that

motivate it change or when circumstances make it advisable.
f) To be:
Clear; concrete and well expressed ideas.
Objective: that defines verifiable characteristics.
Precise: avoid ambiguities.

Effective, it solves the problem that caused it, in a practical and
complete way.
APPLICATION LEVELS
Company Standards.- Those that regulate and control production,

specifying in detail the way in which products are made and oriented,
within the company in its conventional relationships, knowing what raw
materials to use, what equipment to use or acquire, what labor. work to
employ, etc.
Association Standards.- Those developed by a group of people or

companies that carry out activities in a field of common interest,
example: ASTM.
National Standards.- They are those established by the competent

National Standardization Organization, existing in each country, in our
case INDECOPI.
International Standards.- These are those established by regional

and/or international standardization organizations. For example: ISO,
COPANT, CEL.
ISO 9000 STANDARDS
The ISO 9000 series is a set of standards aimed at organizing company

management that have gained international recognition and acceptance
due to the greater power that consumers have and the high international
competition accentuated by integrationist processes.
Some of these standards specify requirements for quality systems (ISO

9001, 9002, 9003) and others give a guide to help in the
interpretation and implementation of the quality system (ISO 9000-2, ISO
9004-1)

5 7
4
ISO 9000 objectives
• Provide elements so that an organization can achieve the quality of

the product or service, while maintaining it over time, so that
customer needs are permanently satisfied, allowing the company
to reduce quality costs, increase productivity, and stand out. or
stand out against the competition.
• Provide clients or users with the assurance that the product or

services have the desired agreed, agreed upon or contracted
quality.
• Provide company management with the assurance that the desired

quality is obtained.
• Establish guidelines by which the organization can select and use

standards.
ISO 9000 standards
The international standardization organization ISO, through its TC sector

- 176 has prepared a set of standards on Management and
Quality Assurance, grouping them by codes from 9000 to 9004, like this:
ISO - 9000: Standards for Quality Management and Quality Assurance,
Guidelines for selection and use.
ISO - 9001: Quality System, quality assurance model in design, development,
production, installation and post service sale.
ISO - 9002: Quality system, quality assurance model in production and
installation.
ISO - 9003: Quality system, quality assurance model in inspection and final
testing.
ISO - 9004: Quality management and elements of the quality system.
Directives.
STANDARD IS014000
ISO 14000 is an international voluntary environmental standard
recognized by major trading nations and trade regulating organizations
such as GATT and the World Trade Organization. It is not a law in the
sense that no one is required to be registered (it is voluntary); However,
no one has to do business with you, buy your products and services, or
allow your products and services into your country if they have declared
ISO 14000 registration a requirement for doing business with them or in
your country. It is expected that many foreign trading partners will require
registration by import manufacturers. This is a legal trade barrier
recognized under international treaty. Elements of the American
Government have indicated intention to institute any preference for, or
requirement that, suppliers be registered. It is likely that registration will
influence the enforcement position of environmental regulators, and will
likely influence insurance rates and lender practices.
ISO 14000 is actually a series of standards covering everything from
environmental management systems (EMS) to auditor qualifications to
as yet unwritten standards for such things as life cycle assessment.
5 7
5
The ISO 14000 standard is not a single standard, but is part of a family
of standards that refer to environmental management applied to the
company, whose objective is the standardization of ways of producing
and providing services that protect the environment. environment,
increasing the quality of the product and, as a consequence, its
competitiveness in the face of the demand for products whose
components and manufacturing processes are carried out in a context
where the environment is respected.
IS014000 Standards
Environmental Management Systems (14001 Specifications and
directives for use - 14004 General directives on principles, systems and
supporting techniques.)
Environmental Audits (14010 General principles. 14011 Audit
Procedures, Environmental Management Systems Audits-14012 Criteria
for certification of auditors)
Environmental performance evaluation (14031 Guidelines - 14032
Examples of Environmental Performance Evaluation)
Life cycle analysis (14040 Principles and general framework - 14041
Definition of the objective and scope and analysis of the inventory.
14042 Life Cycle Impact Assessment - 14043 Life Cycle Interpretation -
14047 Examples of the application of ISO 14042 - 14048 Format
analysis data documentation)
Environmental labels (14020 General principles - 14021Type II - 14024
Type I - 14025 Type III)
Terms and definitions (14050 Vocabulary)
CTION MECHANICS , 52/147

5 7
6
ABSOLUTE, MEAN, MODE, MEDIAN FREQUENCY DISTRIBUTION
Frequency distribution table
If the data available are numerous, it is essential to classify them in a

table or summary table of the original observations, which from now on
we will call the Frequency Distribution Table or simply “Frequency
Table”.
1. Discrete Variable Data

Let i , x , ...,
x 2
Definition: The absolute frequency of the value y is called the number of

times this value appears in the set of observations and is represented by
“n”, i = 1,2,...,m.
Table 1. Absolute Frequency Distribution
Different values observed Absolute frequencies

and ni
and neither
and . n2.
. .
ym nm
Totals n
A survey is carried out on a group of 16 families about the number of

children. The results in the order obtained are:
x i = 2, X 2 = 1, X3 = 3, X4 = 1, X5 = 2,
Xe = 2, X io = 1, X ii = 2, X 12 = 3, X 13 = 4, X 14 = 1,
a. Construct an absolute frequency distribution table for this data.

b. Interpret each absolute frequency (n i )
Solution:
1. First, we observe that the set of observations is of size 16, then n = 16
2. Second, we identify the number of different observations, m = 5.

These values are:
y i = 0, y 2 =1, ya = 2, y 4 = 3, ys = 4

5 7
7
Which are located (in that order) in the first column of table 2.
3. Let us now count the number of families with 0,1,2,3 and 4 children
respectively. This can be done by adding a column in table 2, called
“count”, using the “sticks” method, which consists of putting a vertical
line or check mark (/) every time the value in question appears,
highlighting every five units in order to facilitate the final calculation
(////). Finally, the number of trays is counted, obtaining the absolute
frequencies (n). In our case.
n i = 1, n2 = 6, na = 5, n4 = 3, ns = 1
In this way, the absolute frequency distribution table of the number of
children per family has been constructed.
Table 2. Frequency Distribution of the number of children per family
Number of children Count
Absolute frequencies n i
and
0 / 1
1 N/ / 6
2 N 5
3 /// 3
4 / 1
Totals 16
b. Interpretation of the ni:
n i = 1, means that in the families surveyed, one does not have children.
n2 = 6, means that in the families surveyed, six have one child each.
So on,
ns = 1, means that in the families surveyed, one has four children.
The frequency distribution table can be constructed not only with respect
to absolute frequencies, but also to other types of frequencies, such as:
relative, accumulated absolute, accumulated relative, etc., which are
defined as follows .
Definition 2: The cumulative absolute frequency “less than”

corresponding to the value y, is called the number of observations less
than or equal to y (x<y).
If we denote this frequency by “Ni”, then:
Ni=n+n2+na + ... + n i
Using summation notation, the cumulative absolute Ni=

frequency is written: ¿nj

5 7
8
Thus, for example 1, we have:
1
Ni = Zn =n=1 j = 1
2
N2 = 2n =n+n2=1 + 6 = 7 j = 1
3
N s = 2n =n+n2+n3=1 + 6 + 5 = 12 j = 1
4
N4 = Zn =n+n2+n+n4=1 + 6 + 5 + 3=15 j = 1
5
N5 = Zn =n+n2+n+n4+ns = 1+ 6 + 5 + 3 + 1=16 j=1
Interpretation: N2 = 7, means that in the survey there are 7 families with

1 or fewer children (or equivalent; in the survey 7 families have been
found with at most 1 child).
N s = 12, means that in the survey 12 families with at least 2 children
were obtained.
Definition 3: the cumulative absolute frequency “greater than”
corresponding to the value y is called the number of observations greater
than or equal to yi (X > y)
TO. . . .r ■ «*,
If this frequency is denoted by N i ”, then:
N i = n i + n i+1 + ... + nm
In summation notation, we write: + m
N*=2n j=1
Thus, for example 2, we have that m = 5, then: =n+n2+na+..+ns=1 +6 +

5 + 3 + 1 = 16
5
N=2n j =
5 =n2+n+.+n=6+5+3 + 1 =15
1
N2=2n
j=1
5 =na+n4+ns = 5 + 3 + 1 = 9
N*=2n, j=1
=n4+n5=3+1=4
N*=2n
5
j
=1
= n 5 =1
N*=2n,
* 5
j=1
* .... . " to . ■■
N 1 = 16, means, the families surveyed have between 0 and 4 children
inclusive;
. .* . . ....................................................................
N 2 = 15, means that in the survey, 15 families have at least 1 child.
. ....................
N 3 = 9, means that in the survey, 9 families have at least 2 children
, .* . . .................................................................... .
N4 = 4, means that in the survey, 4 families have at least 3 children
Ns=1, means that of the respondents, only one family has 4 children.

5 7
9
Definition 4: The relative frequency of the value yi is called the quotient
of the absolute frequency of y and the total number of observations.
This frequency is denoted by “hi”. so
_ Absolute frequency of y i _ neither

h i
= total number of observations = n
For example 2 the relative frequencies are:
h _n
_ 1 = 0.0625 ,_n4 htn _ 3 = 0.1875
=n 16 16
h _n __6_ = 0.3750 . n5 _ 1 = 0.0625
2 =n 16 hesn 16
h _n _5
= 0.3125
s =n 16
Interpretation:
h i = 0.0625, it is interpreted as the proportion of families that do not have

a child in the sample.
h 2 = 0.3750, is interpreted as the proportion of families that have a child
in the sample.
So on.
h 5 = 0.0625, is the proportion of families that have 4 children in the
sample.
The importance of frequency is that it measures the “weight” that each
absolute frequency has with respect to the total number of observations.
Definition 5: The relative frequency h i multiplied by 100% (100hi%) is
called percentage relative frequency, and represents the percentage of
observations that corresponds to the value y.
The percentage frequency will logically be between 0 and 100 and is
denoted by:
p i = 100h i Thus, for example 2 we have:
p i= 100h i % = 6.25%, percentage of families that do not have
children in the sample size.
p 2 = 100h 2 % = 37.5%, percentage of families with 1 children in the

sample.
So on:
ps = 100h 5 % = 6.25%, percentage of families with 4 children in the

study sample.
Definition 6: The cumulative relative frequency “Less than” of the

value y is called the total relative frequency of observations less than or
equal to y.
If we denote this frequency by “H”, then

Hi=h1+h2 .+hi= n1+n2+.. + n i

6 7
0
+. n
Using summation notation, H i is written
H= 2
For example 2, the H i are:
1
H=2h=h= 0.0625 j = 1 2
H2=2h=h+h2= 0.0625 + 0.3750 = 0.4375 j = 1
3
H3=2h=h+h2+ha= 0.0625 + 0.3750 + 0.3125 = 0.7500 j = 1 4

H4=2h=h+h2+h+h.= 0.0625 + 0.3750 + 0.3125 + 0.1875 = 0.9375 j=1
5
Hs=2h,=h+h2+h3+h4+hs= 0.0625 + 0.3750 + 0.3125 + 0.1875
j
+ 0.0625= 1.0
Interpretation:
H 2 = 0.4375, is the proportion of families under study with at most 1
child.
H 3 = 0.75, the proportion of families under study with at most two
children.
So on.
Definition 7: The cumulative relative frequency “Greater than” of the
value y is called the total relative frequency of observations greater than
or equal to y
If we denote this frequency by “H”, then
m
H i = 2hj=h+ h i + 1 + ... + h m
For example 2, the H i are:

5
H 1 = 2 h j =h 1 +h 2 + h 3 + h 4 + h 5 =1.0
j=1
5
H2 = 2 h j = h 2 + h 3 + h 4 + h 5 = 0.3750 + 0.3125 + 0.1875 + 0.0625 =

0.9375
j=2
5
H3 = 2 h j = h 3 + h 4 + h 5 = 0.3125 + 0.1875 + 0.0625 = 0.5625

j=3
H4=2h,=h4+h5= 0.1875 + 0.0625 = 0.25 1=4

5
H5=Zh,=hs= 0.0625 í = 5
Interpretation:
H2 = 0.9375, is the proportion of families in the sample with at least 1

child.
Ha = 0.5625, is the proportion of families in the sample with at least two

children.
So on.
Definition 8: The cumulative relative frequency H(H*) multiplied by
100% is called the percentage cumulative relative frequency. That is,
100Hi% (100 H ¡ * %) and represents the percentage of observations less
than or equal to y (greater than or equal to y)

6 7
1
So, for example 2:
100 H2% = 43.75%, is the percentage of families in the sample with at

most 1 child;
100 H3 % = 75%, is the percentage of families in the sample with at most

children;
100 H2 % = 93.75%, is the percentage of families in the sample with at

least 1 child;
100 H3 % = 56.25%, is the percentage of families in the sample with at

least 2 children;
Therefore; The different frequency distribution tables that can be

prepared for the same data set are summarized in Table 3.
Table 3. Different types of Frequency Distribution tables

*
and neither Neither hi hi
N; H* 100 h,% 100H,% 100 H %
i
and. n, N, h, H, N; H; 100 100H, 100H *

and n 2 N 2 h 2 H 2 N 2 * H* hours,
100 100 hours 100H *
hours 2 2
Hm
ym nm N =n
m hm H m n: * 100 h m 100 H m 100H;
m
m
Totals 2n,=n 2h,=n
i=1
100 %
i=1
The different variants of frequency distribution tables for example 2

Table 4. Different types of Frequency Distribution tables of the number of children
and neither Neither hi hi N' H' 100 h,% 100 h% *

100 H ' %
0 1 1 0.0625 0.0625 16 1.00 6.25 6.25 100.00
Yo 6 7 0.3750 0.4375 15 0.9375 37.50 43.75 93.75
2 5 12 0.3125 0.7500 9 0.5625 31.25 75.00 56.25
3 3 15 0.1875 0.9375 4 0.2500 18.75 93.75 25.00
4 1 16 0.0625 1.00 1 0.0625 6.25 100.00 6.25
Totals 16 1.0 100%
Properties of frequencies
What follows are trivial properties that are consequences of the
definitions of absolute, relative, cumulative absolute, and cumulative
frequency, so none of them will be proven.
Property 1: Absolute frequencies and cumulative frequencies are
always non-negative integers. That is to say
. . to. to.* - ..
N ¡ > 0,N ¡ > OyN ¡ > 0, for i=1,2,.,m

Property 2: The sum of all absolute frequencies is equal to the total
number of the set of observations. I mean
m
> ni = ni + n 2 +..+nm = n
Property 3: Relative frequencies and cumulative relative frequencies are

6 7
2
always non-negative fractional numbers, not greater than one. That is to
say.
. . . . .* ...
0 < h i < 1,0 < H i < i and H i < i, for i=1,2,.,m
Property 4: The accumulated absolute frequency “less than”
corresponding to the value ym coincides with the total number of the set
of observations. that is to say:
m
Nm = 2 nis n
Property 5: The cumulative absolute frequency “greater than”

corresponding to the value y matches the total number of the data set.
That is to say
m
Ni= X n=n
Property 6: The cumulative relative frequency “less than” corresponding

to the value ym coincides with unity. That is to say
m
Hm=2hi=1

6 7
3
Property 7: The cumulative relative frequency “greater than”
corresponding to the value y, coincides with unity. I mean:
m
H * = 2 h i =i
Property 8: The absolute frequency accumulated “less than”

corresponding to value y i is n i . That is to
say:
Yo
Ni=2ni=ni
Property 9: The absolute frequency accumulated “greater than”

corresponding to the value ym is n m . I
mean:
..•_ M
N m = 2 ni=nm
i=m
Property 10: The relative frequency accumulated “less than”

corresponding to value y i is h i . That is to
say:
Yo
Hi= 2 hi= ne=i
Property 11: The relative frequency accumulated “greater than”
corresponding to the value ym is h m . I
mean:
,,* m, 1
H m ^^ h i — h m
i—m
Property 12: The accumulated absolute frequencies satisfy the following

inequalities:
n Ni<N2<..<Nm=nyNi= Ni-i + n i
For i = 1, 2,... ,m,y N o =0 Note that n i = + Nn
n =N, > N2 > ... > N =n m and N i * = ND - nufor i = 2,3, ...m

Note that n M = NM - N i
Property 13: The cumulative relative frequencies satisfy the following
inequalities:
— H i < H 2 < ... < H m — i andH i — H ii + h i For i — i,

hi ,m
H o 2,...
— 0 Note that h i — H i + Hu
1=Hi > H2 > ... > H m * =h m and H i * = N ii * - n. for i = 2.3,...m

AI to AI*AI*
Note that n M = NM - N i
Knowledge of these properties serves to control the correct preparation

of frequency tables.

6 7
4
Some data are missing from the frequency table below. Complete it.
and n h, N H
, 2
0
1 5
2 9
3 14
4 0.2 0.70
Solution:
1. Bythe property 8, Ni = n i , so n i = 2
2. Bythe property 12, N 2 = N i + n 2 = 2 + n 2 = 5, from which n2=3
Na = N 2 + na = 5 + na = 9, then na = 4
N4 = N + 3 n4 = 9 + n4 = 14, therefore n4
= 5
3. By the definition of cumulative relative frequency LESS THAN,

n + n2 + n3 + n4 14
H4=h1+h2+hs + h4= ---n---------- =m = 0.70, then n = 20
4. By definition of relative frequency H5 ne ne

n5=4 n =20 = 0.20, from which
5. Finally N5 =N4+ns = 14 + 4 = 18 Property 12
N e =N 5 + n =18 + 4 = 18
6 Property12y4
Then, n e =2
6. The missing relative frequencies are:

7 .3 4 5
h1=20=0.10 h2=20=0.15 h3=20=0.20 = 0.25h4= 20
2
he
20 = 0.10
=
7. The missing cumulative relative frequencies LESS THAN are:
H 1 =h 1 = 0.10 H2=h +h 2 = 0.10 + 0.15 = 0.25 Property10y13
H 3 =h 2 +h 3 = 0.25 + 0.2 = 0.45
H 5 =h 4 +h 5 = 0.70 + 0.20 = 0.90
H e =h 5 +h e = 0.90 + 0.10 = 1.0
Therefore the complete frequency distribution table is as follows
Table 5
and n h, N H
0 2 0.10 2 0.10
1 3 0.15 5 0.25
2 4 0.20 9 0.45
3 5 0.25 14 0.70
4 4 0.20 18 0.90
5 2 0.10 20 1.00
Totals 20 1.00
An agronomist visited 25 orange agricultural cooperatives in the Huaral
valley and in each one he recorded the number of plants attacked by a
certain fungus, from which the following data resulted.

6 7
5
15, 20, 25, 15, 18,16, 17, 18, 20, 18, 18, 18, 19, 16, 17, 19, 16, 17, 19,
16, 17,17,17,19,18,19,18,15
a. Tell what type of data this is.
b. Construct a frequency distribution table appropriate to this set of
values.
c. How many agricultural cooperatives have at most 20 plants attacked
by fungi?
d. How many agricultural cooperatives have at least 17 plants attacked
by fungi?
e. What proportion of cooperatives under study have 18 plants attacked
by fungi?
f. What proportion of cooperatives have 18 or fewer plants attacked by
fungi?
g. What percentage of cooperatives have 18 or fewer plants attacked by
fungi?
Solution
a. Since the data are obtained as a result of counting plants attacked by

fungi, they are discrete.
b. 1. The set of observations is of size 25, so n = 25

a. The different values of the observed characteristic are:
y 1 = 15,y 2 =16,y s =17,y =18,y 5 =19,y 6 = 20, and z = 25

4
That is, m = 7
b.Counting: the number of cooperatives with 15, 16, 17, 18, 19, 20
and 25 plants attacked by fungi respectively is counted. Thus, the
number 15 appears 3 times in the data set, so n 1 = 3; the number
16 also appears 3 times, so n2 = 3; the number 17 appears 5 times,
then na = 5; 18 appears 7 times, that is, na = 7; 19 appears 4 times,
that is; ns = 4; 20 appears twice, then ne = 2; and finally 25 appears
once, that is, n7 = 1
c. The relative frequency of y ¡ , i-1,2, ..., 7, are:
h1 n 1 3 h 5 ns. = =
= 0,12 n" 25 •
= n ' ' 25 ' = 0,16
n 2_ 3 h6 n 6 . 2 =
h2 = = 0,12
n ' ' 25 ' = n" ' 25 ' 0,08
h3 n 3 . h 7 n 7. 1 =
= 0,20
= n' = 25 = n" ' 25 ' 0,04
h 4 n4.
= n' -Z
= 0,28

6 7
6
s. The accumulated absolute frequencies for y ¡ , ¡=1,2,...7, are given by:

Ni=n:=3 Ns=N4+ns = 18 + 4 = 22
na = N i +na= 3 + Ne = N5 + = 22 + 2 = 24
3 =6 N 7 = n = 25
na =N2+ = 6 + 5 = 11
N,=N+n=11+7=18
N ; =n = 25 N J =N J -n 4 =14-7 = 7
N2=Nn=25-3=22 N=Nn=7-4=3
N J =N ; -n 2 = 22-3=19 N*=N*-n=3-2=n=1
N ; =N ; -n 3 =19-5 = 14
b. The accumulated relative frequencies for y ¡ , i=1,2,...7, are:
H:=h=3 HJ = 1
H2=H + h = 0.12 0.12 = 0.24
2
H2=Hh: = 1 -0.12 = 0.88 H ; = H J -
H3 = H 2 + = 0.24 0.20 = 0.44 h = 0.88-0.12 = 0.76 HH J -h 3 =
2
H = + = 0.44 H 5 0.28 = 0.72 0.76- 0.20 = 0.56 H J = H J -h 4 =

=H 4 + h 5 = 0.72 H 0.16 = 0.88 0.56- 0.28 = 0.28 H j =H ; -h 5 =
6 =H 5 + h 6 = 0.88 0.08 = 0.96 0.28-0.16 = 0.12 H J = H J -h 6 =
H 7 =H 6 + h 7 = 0.04 = 1 0.12 - 0.08 = 0.04 = h 7
0.96
Therefore, the different types of frequency distribution tables that can be
prepared for the same set of data are given in Table 6.
Table 6. Frequency Distribution of the number of plants attacked by the fungus
Different Relative Cumulative Cumulative Cumulative Cumulative
Absolute
values frequency absolute absolute relative relative
frecuency
observed frequencies frequencies frequencies frequencies
less than greater than N less than greater than
and. neither hi Neither
; hi H.J.
15 3 0.12 3 25 0.12 1.00

16 3 0.12 6 22 0.24 0.88
17 5 0.20 11 19 0.44 0.76
18 7 0.28 18 14 0.72 0.56
19 4 0.16 22 7 0.88 0.28
20 2 0.08 24 3 0.96 0.12
25 1 0.04 25 1 1.00 0.04
Totals 2n = 25 2h i =1
c. The number of cooperatives that have at least 20 plants attacked by

fungi is N6= 24
d. The number of cooperatives that have at least 17 plants attacked by
fungi is N= 19
e. The proportion of cooperatives in the sample that have 18 plants
attacked by fungi is ha = 0.28, which expressed as a percentage is
28%.
f. The proportion of cooperatives that have 18 or fewer plants attacked
by fungi is H4 = 0.72
g. El100H 4 % = 72%
Note: We will briefly discuss the usefulness of constructing a frequency
distribution table. It is clear that with our examples, where we have only
considered a few data due to the problem of simplifying the calculations,

6 7
7
it is possible that this procedure fully shows all its advantages; However,
with a little imagination on the part of the reader, the experiences
derived from these examples can easily be extended to another situation
where the number of data is greater. In this way, from table 6, which
provides the different variants of frequency tables that can be
constructed for example 3, apart from the questions already answered, it
is possible to extract, for example, conclusions such as the following:
1. In general, there are no major differences in the numbers of plants

attacked by fungi in the 25 cooperatives. The number of attacked plants
that is most repeated is 18, since there are 7 cooperatives that have this
figure (it is given by the absolute frequency of the value 18).
2. If we have criteria on the number of plants attacked with fungi that

each cooperative must have to be declared in emergency by the Ministry
of Agriculture, let's say that this number was 100, from the data in the
table it can be seen that all the cooperatives are below this number,
therefore they will not be declared an emergency.
Observation: the reader will observe that the construction of frequency
tables for discrete data is nothing more than a procedure by which the
information contained in the original set has been considered. We say
this because from the original set of n observations we have gone on to
consider a smaller set composed only of m different values (m < n) (yi).
Now, this procedure that leads to the construction of a frequency table
does not imply loss of information contained in the set of original
observations, since from the knowledge of the yi, i=1,2,...m and their
respective absolute frequencies, the original set of values of these
observations can be constructed without taking into account the order of
their appearance. That is, given such a frequency table, we can know
the number of times each different value must appear in the original set
and thus enumerate each of them; If the order of appearance is not
important, then it can be ensured that the original set of observations is
completely known.
Continuous variable data
We will also designate the original set of data or observations by x1,

x2,..., Xn, with n being the size of the set (sample size) and where these
data are of the continuous type. In this case, because the magnitude of
the characteristic can take, at least theoretically, an infinite number of
values, the process of reduction, grouping or condensation of the
original data, which leads to the construction of frequency tables, is not
so simple. as in the case of discrete data and we cannot proceed as in
the case already studied; It is rather a data classification problem where
the subjectivity of man has an influence that should not be ignored.

6 7
8
Definition 9: Each of the groups into which the data set is divided is
called classes .
The first step to take to solve this classification problem is to say which
and how many classes should be considered. To do this, we normally
start by determining the observation that has the maximum value "Xme",
and the observation that has the minimum value "Ximin"
X max = max {x/i=1,2,., N}, Xmin = min {x/i=1,2,., n}
These extreme values define the range or range of the set of
observations, which is given by the interval [X min ,Xmex
--------------1----------------------1--------------------R
Xm Xmax
in Definition 10: Traversal Width (I) is the
traversal length of the data set. I mean.
/ — Xmin - Xmax
Once the number of classes into which the data will be divided is
determined, let's say m, which can be done with a certain degree of
arbitrariness, since it depends on the problem in question and ultimately
the researcher or statistician, the path of the set of data is divided. data
in as many intervals as desired class, which from now on we will call
class intervals (or simply class). In practice, it is usually divided not
exactly the route [X min, Xmak] but a minimum interval [Xrin , be a multiple of the
number of classes m
[x min ,Xmax ] C [x min ,x max ]
Notation: Let m be the number of classes and, y'i, y'2, ... , y'm be the
ends or limits of the class intervals, then.
y'-y' ,i — 1,2,...,m
Denotes the ith class interval. Where y'-1, is the lower end, and' i the
upper end of the interval.
A class interval, which at least theoretically has no lower or upper limit is
called an open class interval.
Definition 11: Class breadth is the length of the interval that defines
the class, i.e. ______________________
c¡ - y' - y' ii ,i — 1,2,...,m
So c - y' i - y' o Is the length of the first class interval
c - y' 2 - y' i Is the length of the second class interval
2
If the intervals have equal length, let's say “c”, that is

cy' i - y'n for all i — 1,2,...,m
So c — l/m
Construction of class intervals
The class intervals are constructed starting from xmin (or x'min), then
the lower and upper class limits (y' 1-1 - y'i) are consecutively assigned

6 7
9
to the classes, as follows.
Lower limit Upper limit

First class interval: y' or X min and 1 X min +cy 0+c
Second class interval: y' 1 X min +c and 2 X min + 2c = y' i + c
Applied Statistics 1
Introduction 2
INTRODUCTION TO STATISTICS 8
2. Division of statistics 8
2.1 Descriptive statistics 8
85+90+ 93+82+95 = 415 = 89 points 9
2.2 Inferential statistics 9
Issues: 11
Solution: 15
Solution: 16
Review exercises related to basic descriptive statistics 16
1. Data collection 18
A)Some procedures and methods for collecting data 18
Disadvantages of interrogation 19
Interrogation methods 19
Frequency for data collection 20
2. Identification of variables 21
Example: 22
Solution: 23
PRACTICE N° 2 25
INTRODUCTION TO STATISTICS APPLIED TO PROCESSES
FOR QUALITY IMPROVEMENT 26
Statistical Process Control: X and R Charts 27
Basic philosophy of quality 32
1.1 Introduction 32
1.2 Concept and evolution of quality 33
Concept: 33
Quality Evolution 34
1.3 Understanding the meaning of total quality 35
1.4 Strategic importance of total quality 37
1.5 Total quality as a new business management system 38
1.6 Reflections for the reader 38
The improvement process towards total quality 38
2.1 Basic principles for achieving total quality 38

7 7
0
3. The production process is throughout the organization. 39

5. The supplier is part of our process. 39
6. Internal supplier-customer chains are essential. 39
7. Quality is achieved by people and for people. 39
8. Establish the zero defect mentality. 39
continuous improvement. 40
consciousness). 40
ll.Quality 40
12. It requires a new culture. 40
2.2 Improvement Modalities 40
a) Continuous improvement 40
b) Radical innovation or improvement 40
2.3 The control cycle for improvement 41
2.4 Activities to start a process towards total quality 41
PHASE I: Decision making 42
PHASE II: Stage preparation and promotion 42
PHASE III: Implementation of improvement processes 42
PHASE IV: Consolidation and interfunctional optimization 43
QA 43
QA 44
1. Functions and principles of quality control 44
1. "Control is not classification or selection" 45
characteristic of the product." 45
3. "The production team is responsible for quality and control" 45
reasons to study them" 45
5. Decisions must be made based on real data." 45
allows analysis." 45
QA 47
Why produce quality products? (Fig. 2) 47
Introduction to standardization 50
1. Standardization 51
General concepts 51
Definition according to ISO 52

7 7
1
Technical rules 53
Types of technical standards 53
APPLICATION LEVELS 54
ISO 9000 STANDARDS 54
ISO 9000 objectives 55
ISO 9000 standards 55
STANDARD IS014000 55
IS014000 Standards 56
Terms and definitions (14050 Vocabulary) 56
Frequency distribution table 57
1. Discrete Variable Data 57
Solution: 57
Ni=n+n2+na + ... + n i 58
Interpretation: 60
Properties of frequencies 62
Solution 66
Continuous variable data 68
Construction of class intervals 69
Lower limit Upper limit 70
Determination of absolute frequency. 75
Observations: 76
Y1-y'i=<y'i-1 , y'i] 77
General rule for developing frequency distributions 77
and. — 93.5494.5 — 94.0 79
Definition 20: Percent cumulative relative frequency at 82
Solution to: 83
+68 97
PRACTICE N°3 103
Time elapsed from receipt of order to delivery (in days) 104
AVERAGE 106
Observation 106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0 106
2x - X|20 107
Example 107
Solution 107
ge1/gn240_24 107

7 7
2
Proposition (Konig) 107

> 2(xx)2 114
Observation 114
Abbreviated calculation 115
The harmonic mean 115
The mean square 121
THE MEDIAN 122
Observation 122
Observation 122
FASHION 124
Observation 125
EXAMPLE 126
PARTIAL EVALUATION I 127
RANGE, VARIANCE, STANDARD DEVIATION 144
MEASURES OF DISPERSION 144
= (0,010)= 0,0320 145
Interpretation and application 145
Breakdown 146
Where: 146
Example 146
X=1 2x 147
a = 12(x-7)2 147
°\4 147
°s\4 147
PRACTICE N° 4 148
from $10, from 148
FREQUENCY HISTOGRAM 149
C ¡ xn =ni(ocxh= h) C ¡ Ci 149
Frequency polygons 151
Warhead cumulative frequency polygons 152
Note: Percentage warhead 152
Solution: 155
Line graphs 158
PRACTICE N° 5 159
1. Concept: 165
2. Application: 165
4. Distribution function 166

7 7
3
5. Typing 167
_x-u 167
- 167
— —- z 167
Z
= npq is N(0,1) 168
Characteristic of the standard (reduced, standard) normal distribution
168
6. Table management, most frequent cases 169
Example: 171
Concept: 173
How to interpret a cause-effect diagram: 173
Examples of cause-effect 173
The seven tools of quality 174
H1 - Cause-Effect Diagrams 174
Exercise: Cause - Effect Diagram 180
DISPERSION DIAGRAM 180
LINEAR CORRELATION 182
SPEARMAN rank correlation coefficient 184
>62d 185
Yule “Q” correlation coefficient. 185
-
138 -:994 185
V(a+b)(c+d)(a+c)(b+d) 186
Correlation and Causality 186
SIMPLE LINEAR REGRESSION 186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0 188
5b 188
2 and xi = b 2x2+ aZx 188
YsYb(X- x) 188
Concept 188
What is it? 189
When it's used? 189
How is it used? 190
Relationship with other tools 191
Application example 191
PRACTICE N° 7 195
Exercise: Pareto Diagram 195
Organized: 195

7 7
4
Work stoppage 195

Getting relative percentage 195
Construction of the diagram and determination of “vital few” 195
Concept: 195
Control elements. 196
2. Control sheet 196
Scatter plots 197
A control chart shows 198
Need for full participation 198
PARTIAL EVALUATION II 200
THE QUALITY ROUTE 205
First Step: Defining the Problem. 205
Second Step: Recognition of the Characteristics of the Problem
(Observation) Activities 206
Third Step: Search for the Main Causes (Analysis) Activities 207
Fourth Step: Actions to eliminate the causes (Action) Activities 207
Fifth Step: Confirmation of the effectiveness of the action
(Verification). Activities 207
Step Six: Permanent elimination of the causes of the problem
(Standardization) 207
Activities: 207
Step Seven: Review of activities and planning of future work
(Conclusions). 208
Activities 208
■
mth class interval: y' m-1 — X min + (m-1)cy m — X min + mc —
Definition 12: The class mark is the midpoint of the interval that
represents said class. It is denoted by “y”, i—1,2,..., M. That is to say.
y— yn 1 +y i ,i — 1,2,...,m
2
Determination of absolute frequency.

Finally, we proceed to determine the number of observations (absolute
frequency n i ) of the original set (X 1 , X 2 ,..., X n ), which belong to each
class interval. This process is known as the counting operation. It
consists of taking each of the recorded observations and assigning a
“card” to the class to which it belongs (“box” method).
The number of tiles is the absolute class frequency. The concept of

absolute frequency is the same as in definition 1; only in this case they
refer to the classes and not the y values.
Table 7 Frequency distribution

7 7
5
Class interval y' i-1 - y' Count Absolute frequency of

class mark and
i class n i
Iand' and n
y'-yz and 2 n2
■ ■ ■
■ ■ ■
and m-1 - and m ym nm
Totals n
Observations:
In special situations it may be possible to have different amplitude class

intervals.
This depends on the problem in question and the researcher's criteria.
Since the theory for tables with different class amplitudes is the same as
the theory with equal amplitudes.
When it is convenient to expand the width of the path in such a way that
it becomes a “manageable” number in future calculations, this extension
should be done as much as possible symmetrically, that is, to the left of
Xmin and to the right of Xmax.
Regarding the number of classes (m), there is no unanimous consensus

among the authors. Bradfor Hill points out that a number of classes
should be taken between 10 and 20, while Camel indicates that the
appropriate number is between 8 and 15. However, common sense tells
us that the number of classes should not be few because the
condensation would be great and with this there would be losses of
strong information in relation to that contained in the original data set.
On the other hand, an excessive number of classes, although it
produces little loss of information, does not reduce or simplify the work,
and sometimes working with the original data itself is preferable. Some
may also be empty, that is, with zero frequency.
We will give below some frequently used rules:
TO. Take the number of classes (m) equal to the nearest integer to 2Vn,
with n being the number of data.
b. m > Vn, where n is the number of observations
c. Sturges' rule: m = 1 + 3.3 log(n), with n the number of observations.
In general, it is recommended that the number of classes be between

five and twenty.
5 < m < 20
The reader will have noticed that the upper limit of a class coincides with
the lower limit of the next class, so when classifying an observation that
coincides with the common class limit, one is faced with the dilemma,
which class should be assigned? assign this value? To avoid this
difficulty, one of the following alternatives is usually taken:

7 7
6
A. Define the ends of the class intervals with one decimal place more
than those contained in the observations.
B. Consider class intervals as intervals of real numbers open on the left

and closed on the right. Thus, the ith interval y' ¡-i - yi that defines the ith
class equivalent to the interval < y'-, y'i], that is
Y1-y'i=<y'i-1 , y'i]
With this alternative, each class excludes the observation that coincides
with its upper limit.
Note: It should be noted that the above is pure notation and does not
affect the class breadth or class marking.
Convention: in the development of this text for the sake of uniformity,

we will assume the third alternative. Obviously other alternatives can be
taken, this is at the discretion of the specialist.
General rule for developing frequency distributions
1. Decide the number m of class to consider, according to observation 4.
2. Determine the path or range of the data and the amplitude (l) of the
path.
3. Determine the width or length of the class intervals. If possible, it is

convenient that the class intervals be of equal amplitude (c), which is
given by
c = l/m
If this is possible, use class intervals of different sizes or open class
intervals.
4. Determine class boundaries, so that each observation is

unambiguously classified into a single class.
5. Class marks are determined.
6. Finally, the absolute frequency of each class is found.
Suppose that 50 observations have been recorded regarding the

weights of 50 steel ingots produced by SIDERPERÚ, the sample was
obtained from weekly production and the units are given in kg.
94.3 93.0 95.5 95.3 92.4 94.4 92.8 93.2 93.6 95.5
92.9 93.6 95.7 93.8 94.8 93.9 92.7 91.6 93.6 93.7
94.2 95.7 94.7 94.3 92.7 94.5 96.2 95.4 93.7 91.9
94.7 92.7 95.0 93.0 92.9 93.7 92.7 93.3 94.6 96.4
94.1 93.7 94.2 93.7 94.0 93.9 93.6 94.6 92.3 94.4
a. Classify these observations in a frequency table with 5 classes of
equal amplitude.

7 7
7
b. Interpret each number

Solution to
1. By data; m = 5
2. From the data set Xmax = max {X1,X2,...X50} = 96.4 and the
amplitudes l = 96.4-91.6 = 4.8
48
3. From (1) and (2) the constant class amplitude will be C=m=*55 0.96 a
number like this is difficult to handle, so just like you. We prefer to
work with the closest integer c = 1. For this we expand the data path
symmetrically to the left of Xmin = 91.6 and to the right of

7 7
8
It can be I = 96.5 - 91.5 = 5, where Xmin = 91.5 and Xmax = 96.5
4. We will determine the class limits:
and o — X min — 91.5, y' 3 — y' 2 + c — 93.5 + 1 —

94.5;
y' i — y' o + c — 91.5+1 —92.5
y' — y' 3 + c — 94.5 + 1 —
y'2 — y' 1 + c — 92.5 + 1 — 95.5
93.5
5. The class marks are found:
9 _ 94.5 + 95.5 _
y1 —91.5192.5 — 92.0 Y4—------2------— 95.0
9 _ 95.5 + 96.5 _
y2 — 92.5293.5 — 93.0 and 5 —- -9------— 96.0
and. — 93.5494.5 — 94.0
6. Finally, the absolute frequency of each class is determined, taking

into account that we are considering class intervals open to the right
and closed to the left, that is.
y' i-1 -y' i E[ y' i-1 , y' i >
For example, the value 94.5 will be in the fourth class.
We proceed as follows: we take the first observation 94.3 and look for
the class interval to which it belongs, it is 93.5 - 94.5, then a slot is
assigned at the intersection of the count column and the row of this
interval. The other observation 93.0 is now taken, which belongs to the
interval 92.5 - 93.5, then a slot is assigned at the intersection of the row
of this new interval and the count column. So on until the last
observation is exhausted. By adding the scores, the absolute frequency
of each class is obtained. Table 8 is obtained in this way.
Count
Class interval [y,y] class mark and Absolute frecuency
neither
91.5-92.5 92.0 //// 4

92.5-93.5 93.0 //// NN/ / 11
93.5-94.5 94.0 NNNN 20
94.5-95.5 95.0 NN NN 9
95.5-96.5 96.0 NN / 6
Totals 2 n i = 50
b. Interpretation of the ni:
n 1 — 4, means that in the sample there are 4 ingots weighing 91.5 kg or

more but less than 92.5 kg
n2 — 11, in the sample, 11 ingots weigh 92.5 kg or more, but less than
93.5 kg, etc.

7 7
9
Note: when preparing the frequency table for continuous variable data;
There is always a slight loss of information, which occurs by no longer
considering the individual data, but rather the class mark (or midpoint of
each class) as a representative value of the entire class. Thus, y2 = 93.0
represents all observations greater than or equal to 92.5 and less than
93.5. However, this loss of information is compensated by the gain in
conciseness, simplicity and clarity.
Definition 14: Cumulative absolute frequency “less than” (N) of the

ith class, is the number of observations less than the upper end of the ith
class interval, that is, less than y' i . So
Ni=n:+n2 + ... +ni=2n j=i
So, for the example.
N i = n i = 4, is the number of steel ingots that weigh less than 92.5 kg

N2 = n i + n 2 += 4 + 11, is the number of steel ingots whose weight is
less than 93.5kg
Na = n i + n 2 + na = 4 + 11 + 20 =35, number of steel ingots weighing
less than 94.5 kg
N 4 = n i + n 2 + n3+ n4 = 4 + ii + 20 + 9 = 44, number of steel ingots
weighing less than 95.5 kg.
No. 5 = n = 50
Definition 15: Cumulative absolute frequency “greater than” (N f ) of

the ith class, is the number of observations greater than or equal to the
lower end of the ith class interval, that is, greater than or equal to y'H . So:
Yo
N i = n i + n w +j=i... + n m = 2 n j
So, for the example:
N i = n i + n 2 + ... + ns = n = 50, number of steel ingots weighing at least

9i.5kg
N 2 = n2+ n3+n4+ns=11 + 20 + 9 + 6 = 46, number of steel ingots

weighing at least 92.5 kg
AI* AA TO TO TO - , 11.1
N 3 = n3+ n4+ ns = 20 + 9 + 6 = 35, number of steel ingots weighing at
least 93.5 Kg.
N 4 = n4 + n 5 = 9 + 6 = i5, number of steel ingots weighing at least 94.5

kg
N 5 = n 5 = 6, number of steel ingots weighing at least 95.5 kg
Definition 16: Relative frequency (hi) of the ith class interval, is the
quotient.
h i = —, i i_=ni,1, 2,..., m

8
0
For example 4:
____________ . .................................................
h1 =— = 5 = 0.08 , is the proportion of steel ingots that weigh
less than 92.5 kg.
h2 =n = e1 = 0.22 , is the proportion of ingot that weighs 92.5 kg.
either more but less than 93.5 kg
ha =n = 50 = 0.40 , is the proportion of steel ingots that weigh
93.5 kg. or more but less than 94.5 kg
.=#=8=0.18
hen=8=0.12
Definition 17: Percentage relative frequency at the relative frequency
hi multiplied by 100% (100%) and represents the percentage of
observations that belong to the ith class.
So for example 4
100 h i % = 100(0.08)% = 8%, is the percentage of steel ingots that weigh
less than 92.5 Kg.
100 h 2 % = 100(0.22)% = 22%, is the percentage of steel ingots that
weigh 92.5 kg or more but less than 93.5 kg
100 h 3 % = 100(0.40)% = 40%, is the percentage of steel ingots that
weigh 93.5 kg or more but less than 94.5 kg
Definition 18: Cumulative relative frequency “less than” (Hi) l of the
ith class, is the relative frequency corresponding to the observed values
less than y'i, that is, less than the upper end of the ith class interval . so.
H i =h 1 +h2 + ... + h m = 2 n
j=1
For example 4:
Hi=hi= 0.08, is the proportion of steel ingots that weigh less than 92.5 kg
H2 = h 1 +h2 = 0.08 + 0.22 = 0.30, is the proportion of steel ingots that
weigh less than 93.5 kg
Ha = +h 2 + h 3 = 0.08 + 0.22 + 0.40 = 0.70, is the proportion of steel
ingots that weigh less than 94.5 kg
Definition 19: Cumulative relative frequency “greater than or equal
to” (H) of the ith class, is the total relative frequency corresponding to the
observed values greater than or equal to the lower end of the ith class
interval, that is, greater than or equal to same as y'-, then.
m
Hi=h+ h¡ + i +..+ = 2 hj j=i
For example 4: m
Hi=2hj=1
j=i
H2 = h2+ h + h 4 + h = 0.22 + 0.40 + 0.18 + 0.12 = 0.92

3 5
is the proportion of steel ingots that weigh at least 92.5 kg *

H 3 = ha+ h 4 +hs = 0.40 + 0.18 + 0.12 = 0.70
is the proportion of steel ingots that weigh at least 93.5 kg

8 7
1
Definition 20: Percent cumulative relative frequency at

cumulative relative frequency H ¡ (or H 1 ) multiplied by 100%
That is, 100 H ¡ % (or 100 H ¡ %) and represents the percentage of
observations less than y' (or greater than or equal to y'-)
So, for example 4:
100 H 1 % = 100(0.08)% = 8%, is the percentage of steel ingots that
weigh less than 92.5 kg
100 H2% = 100(0.30)% = 30%, is the percentage of steel ingot that
weighs less than 93.5 kg
Etc.
100H2% = 100(0.92)% = 92%, is the percentage of steel ingots that
weigh at least 92.5 kg
100H3% = 100(0.70)% = 70%, is the percentage of steel ingots that
weigh at least 93.5 kg
Therefore, the different frequency tables that can be prepared for a
set of continuous data are given in Table 9.
Neit
y' i-1 -y' i and neither hi Neither hi
her
hi 100h i % 100H i % 100H 1 %
and 0 - and and n h1 No. H 1 N 1 100h 1 100H 1 100
and 1 - and and n 2 . h 2 .
1 1
No. H Neit H i . 100h 2 . 100H 2 . 100H 1 .
2 . 2 . 2 . her .
2 .
. . . . . . . . . . .
. . . . . . . . . 100
and m-1 - ym nm h m n 1 Nm h i m 100h m 100Hm
Totals n 1 100
The different frequency tables that can be formed for example 4 are
summarized in table 10
Neit
[y' ii - and
neither hi Neither hi
her
hi 100h i % 100H i % 100H i %
91.5- and
92.5 92.0 4 0.08 4 0.08 50 1.00 8 8 100
92.5- 93.5 93.0 11 0.22 15 0.30 46 0.92 22 30 92
93.5- 94.5 94.0 20 0.40 35 0.70 35 0.70 40 70 70
94.5- 95.5 95.0 9 0.18 44 0.88 15 0.30 18 88 30
95.5- 96.5 96.0 6 0.12 50 1.00 6 0.12 12 100 12
Totals 50 1 100
Note: Before moving on, it is worth illustrating, as was done with the
results in Table 6, how valuable and summarized information can be
obtained from a classification of continuous data. In example 2.4 where
the observations regarding the weights of 50 steel ingots produced by
SIDERPERÚ are considered representative OF the entire population of
steel ingots produced by said company, the different frequency
distributions that appear in table 10 can, among others, , provide
information such as the following:
1. That the highest percentage of steel ingots weigh between 93.5 and
94.5 kg. As can be seen in the column corresponding to 100h%, this
percentage is 40.
2. If the appropriate weight of steel ingots for acceptance in the

international market is at least 93.5%, the results in the table reflect

8 7
2
that only 70% (see column 100 Hi %) of this population meets the
condition, which Perhaps it represents that said company must
readjust its machinery, since there would be a 30% that cannot be
done from now on.
The real annual investment (in thousands of dollars) of a group of small
businesses were:
10 12 8 40 6 8 10 30 2 8 6 14
16 20 25 28 30 26 30 4 6 10 18 17
13 17 21 7 6 8 14 715 19 27 22
0 14 6 8 9 11 13 15 18 20 30 60
12 6 5 5 6 8 712 15 36 39 52
It is requested:
a. Form a table of all frequency distributions, with intervals of constant

amplitude;
b. The number of small companies with investments of less than 40
thousand dollars.
c. The number of small businesses that have invested $40,000 or more.
d. The number of small businesses that invested $10,000 or more but
less than $35,000.
e. The proportion of small companies with investments less than 8
thousand dollars.
f. The proportion of small businesses that invested at least 20 thousand
dollars but at most 38 thousand dollars.
g. The percentage of small businesses that have invested between 6
thousand dollars and 30 thousand dollars inclusive;
h. The percentage of small businesses that invested more than 64
thousand dollars.
Solution to:
1. Let us consider m >V60 = 7.746, that is, m = 8
2. Xmx = 60 and Xmin = 0, therefore l = 60 - 0 = 60
3. From (1) and (2) cen=6=7.5

8 7
3
Obviously we prefer to work with c = 8. For which we expand the data

path, so that it is a multiple of m = 8. But since the investment cannot be
x max = 64 and Xmin = X min =

or
Now: / =64-0 = 64
4. The class limits are:
y' o = X' min = 0
y' i = y' o +c = 0 + 8
y' = y' or +2c = 0 + 2x8 =16
y' = y' or +3c = 24
y' 4 = y' o + 4c = 32
y' = y' or +5c = 4o
5. The class marks are:
negative, we expand only to the right, that is, we do:
y' 6 = y' o + 6c =
48
y' = y' o + 7c=56
7
y'e = y'o + 8c=64
or + 8 . , 8 + 16 . TO . - .
y= 2 =4 Y2 =—2— = 12 Note that this is the same as
y2 =y i + c-4 + 8-12. Then y3 -y2 + c-12 + 8-2o y 4 = y 3 + c = 2o + 8 =

28, etc.
b. The absolute frequencies of each class are obtained following the
aforementioned process. The other frequencies will also be located
directly in the table.
Table 11 Frequency distribution of small business investments
y' i- NEIT
1
- y'i
and
neither hi Neither hi
HER h 100h i %
eithe - 8 4 15 15/60 15 15/60 6th 16/60 150/6
8 r - 16 12 22 6/22 37 37/60 45 45/6o 22nd/6
16 -24 2nd 1st 10/60 47 47/6th 23 6/23 100/6
24 - 32 28 8 8/6th 55 55/6th 13 13/6th 8th/6
32 -4o 36 2 2/6th 57 57/6th 5 5/6th 2nd/6
4th -48 44 1 1/60 58 58/60 3 3/6th 10/6
48 - 56 52 1 1/60 59 59/6th 2 2/6th 10/6
56 -64 6th 1 1/60 6th 60/60 1 1/60 10/6
Totals 6th 1 100%
8. The number of small companies with investments of less than 40

thousand dollars is 57. This number is located at the intersection of the

8 7
4
interval row "32-40” and the Ni column. (Fig. 2)
15 + 22 +10 + 8+2 = 57 1+1+1=3
Fig.2
c. The number of small companies with investments of 40 thousand
dollars or more is 3 and is found at the intersection of the row of the
interval “40*48” and the column Ni (Fig. 2)
d. Let's locate the points 10 and 35 and the class intervals on a line (Fig.
3)
Applied Statistics..................................................................................................1
Introduction............................................................................................2
INTRODUCTION TO STATISTICS....................................................8
2. Division of statistics.......................................................................8
2.1 Descriptive statistics...................................................................8
85+90+ 93+82+95 = 415 = 89 points....................................................9
2.2 Inferential statistics.....................................................................9
Issues:...................................................................................................11
Solution:...............................................................................................15
Solution:...............................................................................................16
Review exercises related to basic descriptive statistics.......................16
1. Data collection..............................................................................18
A)Some procedures and methods for collecting data..........................18
Disadvantages of interrogation............................................................19
Interrogation methods..........................................................................19
Frequency for data collection...............................................................20
2. Identification of variables................................................................21
Example:..............................................................................................22
Solution:...............................................................................................23
PRACTICE N° 2..................................................................................25
QUALITY IMPROVEMENT.............................................................26
Statistical Process Control: X and R Charts.........................................27
Basic philosophy of quality.................................................................32
1.1 Introduction...............................................................................32
1.2 Concept and evolution of quality..............................................33
8 7
5
Concept:...............................................................................................33
Quality Evolution.................................................................................34
1.3 Understanding the meaning of total quality..............................35
1.4 Strategic importance of total quality.........................................37
1.5 Total quality as a new business management system...............38
1.6 Reflections for the reader...............................................................38
The improvement process towards total quality..................................38
2.1 Basic principles for achieving total quality...................................38
3. The production process is throughout the organization................39
5. The supplier is part of our process................................................39
6. Internal supplier-customer chains are essential............................39
7. Quality is achieved by people and for people...............................39
8. Establish the zero defect mentality...............................................39
continuous improvement......................................................................40
consciousness)......................................................................................40
ll.Quality..............................................................................................40
12. It requires a new culture............................................................40
2.2 Improvement Modalities...........................................................40
a) Continuous improvement.............................................................40
b) Radical innovation or improvement.............................................40
2.3 The control cycle for improvement...........................................41
2.4 Activities to start a process towards total quality..........................41
PHASE I: Decision making.................................................................42
PHASE II: Stage preparation and promotion.......................................42
PHASE III: Implementation of improvement processes.....................42
PHASE IV: Consolidation and interfunctional optimization...............43
QA........................................................................................................43
QA........................................................................................................44
1. Functions and principles of quality control..................................44
1. "Control is not classification or selection"...................................45
characteristic of the product."..............................................................45
3. "The production team is responsible for quality and control"......45
reasons to study them".........................................................................45
5. Decisions must be made based on real data."...............................45
8 7
6
allows analysis."...................................................................................45
QA........................................................................................................47
Why produce quality products? (Fig. 2)..............................................47
Introduction to standardization............................................................50
1. Standardization.............................................................................51
General concepts..................................................................................51
Definition according to ISO.................................................................52
Technical rules.....................................................................................53
Types of technical standards................................................................53
APPLICATION LEVELS...................................................................54
ISO 9000 STANDARDS.....................................................................54
ISO 9000 objectives.............................................................................55
ISO 9000 standards..............................................................................55
STANDARD IS014000.......................................................................55
IS014000 Standards.............................................................................56
Terms and definitions (14050 Vocabulary).........................................56
Frequency distribution table................................................................57
1. Discrete Variable Data..................................................................57
Solution:...............................................................................................57
Ni=n+n2+na + ... + n i............................................................................58
Interpretation:.......................................................................................60
Properties of frequencies......................................................................62
Solution................................................................................................66
Continuous variable data......................................................................68
Construction of class intervals.............................................................69
Lower limit Upper limit....................................................................70
Determination of absolute frequency...................................................75
Observations:.......................................................................................76
Y1-y'i=<y'i-1 , y'i].................................................................77
General rule for developing frequency distributions...........................77
and. — 93.5494.5 — 94.0.......................................................................79
Definition 20: Percent cumulative relative frequency at...................82
Solution to:...........................................................................................83
+68......................................................................................................................97
PRACTICE N°3.................................................................................103
Time elapsed from receipt of order to delivery (in days)..................104
AVERAGE........................................................................................106

8 7
7
Observation........................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0............................................................106
2x - X|20..............................................................................................107
Example.............................................................................................107
Solution..............................................................................................107
ge1/gn240_24............................................................................................107
Proposition (Konig)...........................................................................107
> 2(xx)2.........................................................................................................114
Observation........................................................................................114
Abbreviated calculation.....................................................................115
The harmonic mean............................................................................115
The mean square................................................................................121
THE MEDIAN...................................................................................122
Observation........................................................................................122
Observation........................................................................................122
FASHION..........................................................................................124
Observation........................................................................................125
EXAMPLE.........................................................................................126
PARTIAL EVALUATION I.............................................................127
RANGE, VARIANCE, STANDARD DEVIATION........................144
MEASURES OF DISPERSION........................................................144
= (0,010)= 0,0320......................................................................145
Interpretation and application............................................................145
Breakdown.........................................................................................146
Where:................................................................................................146
Example.............................................................................................146
X=1 2x..............................................................................................................147
a = 12(x-7)2..........................................................................................147
°\4..................................................................................................................147
°s\4............................................................................................................147
PRACTICE N° 4................................................................................148
from $10, from..............................................................................................148
FREQUENCY HISTOGRAM...........................................................149
C ¡ xn =ni(ocxh= h) C ¡ Ci................................................149
Frequency polygons...........................................................................151
Warhead cumulative frequency polygons..........................................152
Note: Percentage warhead..................................................................152
8 7
8
Solution:.............................................................................................155
Line graphs.........................................................................................158
PRACTICE N° 5..................................................................................159
1. Concept:......................................................................................165
2. Application:................................................................................165
4. Distribution function...................................................................166
5. Typing.........................................................................................167
_x-u.......................................................................................................167
-..............................................................................................................167
— —- z.................................................................................................167
Z
= npq is N(0,1)..........................................................................................168
............................................................................................................168
6. Table management, most frequent cases....................................169
Example:............................................................................................171
Concept:.............................................................................................173
How to interpret a cause-effect diagram:...........................................173
Examples of cause-effect...................................................................173
The seven tools of quality..................................................................174
H1 - Cause-Effect Diagrams..............................................................174
Exercise: Cause - Effect Diagram......................................................180
DISPERSION DIAGRAM................................................................180
LINEAR CORRELATION................................................................182
SPEARMAN rank correlation coefficient.........................................184
> 6 2 d...........................................................................................................185
Yule “Q” correlation coefficient........................................................185
-
138 -:994........................................................................................185
V(a+b)(c+d)(a+c)(b+d).......................................................................186
Correlation and Causality..................................................................186
SIMPLE LINEAR REGRESSION....................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0........................................................188
5b...........................................................................................................188
2 and xi = b 2x2+ aZx.............................................................188
YsYb(X- x)................................................................................................188
Concept..............................................................................................188
What is it?..........................................................................................189
When it's used?..................................................................................189
How is it used?...................................................................................190

8 7
9
Relationship with other tools.............................................................191

Application example..........................................................................191
PRACTICE N° 7................................................................................195
Exercise: Pareto Diagram..................................................................195
Organized:..........................................................................................195
Work stoppage...................................................................................195
Getting relative percentage................................................................195
Construction of the diagram and determination of “vital few”..........195
Concept:.............................................................................................195
Control elements................................................................................196
2. Control sheet..................................................................................196
Scatter plots........................................................................................197
A control chart shows........................................................................198
Need for full participation..................................................................198
PARTIAL EVALUATION II............................................................200
THE QUALITY ROUTE...................................................................205
First Step: Defining the Problem.......................................................205
(Observation) Activities.....................................................................206
Third Step: Search for the Main Causes (Analysis) Activities..........207
Fourth Step: Actions to eliminate the causes (Action) Activities......207
(Verification). Activities....................................................................207
(Standardization)................................................................................207
Activities:...........................................................................................207
(Conclusions).....................................................................................208
Activities............................................................................................208
x
Fig.3
Therefore, the number of small companies that invested 10 thousand

dollars or more but less than 35 thousand dollars will be:
X+ 10 + 8 + y
X, is determined by linearly interpolating as follows: assuming that the
data are uniformly distributed over the class intervals, then
If in an amplitude of 8 = 16 - 8 there are 22 values
In an amplitude of 6 = 16-10 there will be x values
. (16- 10)22 6x22 _
That isx = —1618— = g = 16.5 = 17
9 7
0
Similarly we obtain and:

in an amplitude of 8 = 40 - 32 there are 2 observations
In an amplitude of 3 = 35 - 32 there will be “and”
observations
(35-32)2 3x2 6 .
oh oh 40-32 8 8 1
Therefore, the number will be 17+ 10 + 8+1 =36 companies

e. The proportion of small companies with investments less than 8
thousand dollars is 15/60 =.0.25
f. Locate the limits of the intervals and the points 20 and 38 on a line
(Fig. 4)
Introduction............................................................................................2
85+90+ 93+82+95 = 415 = 89 points....................................................9
Issues:...................................................................................................11
Solution:...............................................................................................15
Solution:...............................................................................................16
1. Data collection..............................................................................18
Example:..............................................................................................22
Solution:...............................................................................................23
PRACTICE N° 2..................................................................................25
QUALITY IMPROVEMENT.............................................................26
Basic philosophy of quality.................................................................32
1.1 Introduction...............................................................................32

9 7
1
Concept:...............................................................................................33
2.1 Basic principles for achieving total quality...................................38
consciousness)......................................................................................40
ll.Quality..............................................................................................40
a) Continuous improvement.............................................................40
PHASE III: Implementation of improvement processes.....................42
QA........................................................................................................43
QA........................................................................................................44
9 7
2
allows analysis."...................................................................................45
QA........................................................................................................47
Technical rules.....................................................................................53
APPLICATION LEVELS...................................................................54
ISO 9000 STANDARDS.....................................................................54
ISO 9000 objectives.............................................................................55
ISO 9000 standards..............................................................................55
STANDARD IS014000.......................................................................55
IS014000 Standards.............................................................................56
Frequency distribution table................................................................57
Solution:...............................................................................................57
Ni=n+n2+na + ... + n i............................................................................58
Interpretation:.......................................................................................60
Solution................................................................................................66
Observations:.......................................................................................76
Y1-y'i=<y'i-1 , y'i].................................................................77
and. — 93.5494.5 — 94.0.......................................................................79
Solution to:...........................................................................................83
+68......................................................................................................................97
PRACTICE N°3.................................................................................103
AVERAGE........................................................................................106

9 7
3
Observation........................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0............................................................106
2x - X|20..............................................................................................107
Example.............................................................................................107
Solution..............................................................................................107
ge1/gn240_24............................................................................................107
Proposition (Konig)...........................................................................107
> 2(xx)2.........................................................................................................114
Observation........................................................................................114
The mean square................................................................................121
THE MEDIAN...................................................................................122
Observation........................................................................................122
Observation........................................................................................122
FASHION..........................................................................................124
Observation........................................................................................125
EXAMPLE.........................................................................................126
= (0,010)= 0,0320......................................................................145
Breakdown.........................................................................................146
Where:................................................................................................146
Example.............................................................................................146
X=1 2x..............................................................................................................147
a = 12(x-7)2..........................................................................................147
°\4..................................................................................................................147
°s\4............................................................................................................147
PRACTICE N° 4................................................................................148
from $10, from..............................................................................................148
C ¡ xn =ni(ocxh= h) C ¡ Ci................................................149
9 7
4
Solution:.............................................................................................155
Line graphs.........................................................................................158
PRACTICE N° 5..................................................................................159
1. Concept:......................................................................................165
2. Application:................................................................................165
5. Typing.........................................................................................167
_x-u.......................................................................................................167
-..............................................................................................................167
— —- z.................................................................................................167
Z
= npq is N(0,1)..........................................................................................168
............................................................................................................168
Example:............................................................................................171
Concept:.............................................................................................173
> 6 2 d...........................................................................................................185
-
138 -:994........................................................................................185
V(a+b)(c+d)(a+c)(b+d).......................................................................186
Correlation and Causality..................................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0........................................................188
5b...........................................................................................................188
2 and xi = b 2x2+ aZx.............................................................188
YsYb(X- x)................................................................................................188
Concept..............................................................................................188
What is it?..........................................................................................189
When it's used?..................................................................................189
How is it used?...................................................................................190

9 7
5

PRACTICE N° 7................................................................................195
Exercise: Pareto Diagram..................................................................195
Organized:..........................................................................................195
Work stoppage...................................................................................195
Concept:.............................................................................................195
2. Control sheet..................................................................................196
Scatter plots........................................................................................197
Activities:...........................................................................................207
(Conclusions).....................................................................................208
Activities............................................................................................208
Fig.4

9 7
6
The proportion of small companies that intervened at least 20

thousand dollars but at most 38 thousand dollars will be given by:
+ +P2
P is determined as follows: assuming that the observations are

uniformly distributed in each class interval, then 6
If in an amplitude of 8 = 24 -16 there is a proportion of 10/60
At an amplitude of 4 = 24 - 20 there will be a proportion p i
Introduction............................................................................................2
85+90+ 93+82+95 = 415 = 89 points....................................................9
Issues:...................................................................................................11
Solution:...............................................................................................15
Solution:...............................................................................................16
1. Data collection..............................................................................18
Example:..............................................................................................22
Solution:...............................................................................................23
PRACTICE N° 2..................................................................................25
FOR QUALITY IMPROVEMENT.....................................................26
Basic philosophy of quality..................................................................32
1.1 Introduction...............................................................................32
Concept:...............................................................................................33
PRODUCTION MECHANICS 97 76/147


2.1 Basic principles for achieving total quality....................................38
consciousness)......................................................................................40
ll.Quality..............................................................................................40
a) Continuous improvement..............................................................40
PHASE III: Implementation of improvement processes......................42
QA........................................................................................................43
QA........................................................................................................44
allows analysis."...................................................................................45

QA........................................................................................................47
Technical rules.....................................................................................53
APPLICATION LEVELS....................................................................54
ISO 9000 STANDARDS.....................................................................54
ISO 9000 objectives.............................................................................55
ISO 9000 standards..............................................................................55
STANDARD IS014000.......................................................................55
IS014000 Standards.............................................................................56
Frequency distribution table.................................................................57
Solution:...............................................................................................57
Ni=n+n2+na + ... + n i............................................................................58
Interpretation:.......................................................................................60
Solution................................................................................................66
Observations:.......................................................................................76
Y1-y'i=<y'i-1 , y'i].................................................................77
and. — 93.5494.5 — 94.0.......................................................................79
Solution to:...........................................................................................83
+68......................................................................................................................97
PRACTICE N°3.................................................................................103

AVERAGE.........................................................................................106
Observation........................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0............................................................106
2x - X|20..............................................................................................107
Example.............................................................................................107
Solution..............................................................................................107
ge1/gn240_24............................................................................................107
Proposition (Konig)............................................................................107
> 2(xx)2.........................................................................................................114
Observation........................................................................................114
The mean square................................................................................121
THE MEDIAN...................................................................................122
Observation........................................................................................122
Observation........................................................................................122
FASHION..........................................................................................124
Observation........................................................................................125
EXAMPLE.........................................................................................126
= (0,010)= 0,0320......................................................................145
Breakdown.........................................................................................146
Where:................................................................................................146
Example.............................................................................................146
X=1 2x..............................................................................................................147
a = 12(x-7)2..........................................................................................147
°\4..................................................................................................................147
°s\4............................................................................................................147
PRACTICE N° 4................................................................................148
from $10, from..............................................................................................148
C ¡ xn =ni(ocxh= h) C ¡ Ci................................................149

Solution:.............................................................................................155
Line graphs.........................................................................................158
PRACTICE N° 5..................................................................................159
1. Concept:......................................................................................165
2. Application:................................................................................165
5. Typing.........................................................................................167
_x-u........................................................................................................167
-..............................................................................................................167
— —- z.................................................................................................167
Z
= npq is N(0,1)..........................................................................................168
............................................................................................................168
Example:............................................................................................171
Concept:.............................................................................................173
> 6 2 d...........................................................................................................185
-
138 -:994........................................................................................185
V(a+b)(c+d)(a+c)(b+d).......................................................................186
Correlation and Causality...................................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0........................................................188
5b...........................................................................................................188
2 and xi = b 2x2+ aZx.............................................................188

YsYb(X- x)................................................................................................188
Concept..............................................................................................188
What is it?..........................................................................................189
When it's used?..................................................................................189
How is it used?...................................................................................190
PRACTICE N° 7................................................................................195
Exercise: Pareto Diagram...................................................................195
Organized:..........................................................................................195
Work stoppage...................................................................................195
Concept:.............................................................................................195
2. Control sheet..................................................................................196
Scatter plots........................................................................................197
Activities:...........................................................................................207
(Conclusions).....................................................................................208
Activities............................................................................................208
5 8 3 ____
Therefore, the requested proportion will be 60+60+ 120 = 0.242

g. The same process of (d) and (f) is followed
150/6% 220/6% 100/6% 80/6%

--------------- >-------------—X
16 40
24 30 32
x% Fig. 5 an
d%
The percentage of small businesses that have invested between 6 and
220
x% + 6
100
6
30 thousand dollars inclusive will be:
X%, is obtained as follows:
If in an amplitude of 8 = 8 - 0, there will be 15/60% In an amplitude of 2 =

8 - 6, there will be x%
(8-6)(150/6)% 2/150)0g
That is: x% = 1 ,—— = o(—e-) % = 6.25% 8-0 8 \ 6 /
In a completely similar way we obtain y%
If in an amplitude of 8 = 32 - 24 there is 80/6%

At an amplitude of 6 = 30 - 24 there will be y% _ , ,
0 (30-24)
(80 / 6)% 6/80
This is y% = * (32 - 24) = 8 (6) % = 10%
PRACTICE N°3
1. Why are relative frequencies more important than absolute

frequencies?
2. When constructing a frequency distribution, the number of classes

used depends on:
A. Data number c. Population size
B. Range of data collected d. All of the above
E.Aybperonoc
3. Explain the difference between relative frequency and percentage

distributions.
4. Below are the ages of 50 members of a government social service

program:
81 53 67 60 80 80 64 56 5491 61
66 88 67 65 52 52 72 74 6573 69
43 54 76 70 97 97 68 82 7579 60

39 87 76 97 86 86 45 60 4565 76
92 72 82 80 70 70 65 50 5870 56
Build relative frequency distributions from this data using 7 and 13

equal intervals. State policies for social service programs require that
approximately 40% of program participants be over 50 years of age.
a. Does the program conform to that policy?
b. Which of the relative frequency distributions helps you best answer

part (a)?
c. Suppose the director of social services wants to know the proportion

of program participants whose age ranges between 45 and 80 years.
With which of the two distributions could the director best estimate the
response?
5. The VELOZ company, a company located in Arequipa, sampled its

shipping records during a certain day, obtaining the following results:
Time elapsed from receipt of order to delivery (in days)

4 12 8 14 11 6 7 13 13 11 11
20 5 19 10 15 24 7 29 6

Construct a frequency distribution for the above data and a relative

frequency distribution. Use 6 day intervals.
a. What statement can you make about the efficiency of order

processing from frequency distributions?
b. If the company wants to ensure that half of its deliveries are made
in 10 or fewer days. Can you determine by frequency distribution
has the company reached its goal?
6. The class marks of a frequency distribution with intervals of equal

amplitude are: 46,55,64,73,82,91. Find:
a. The class interval c. The class limits
b. The range
7. There is a frequency distribution with four class intervals of equal

amplitude and the following data:
and i = 10,and 4 =22 h i = 0.30 ha=17.5% H2=0.45 n = 120
Reconstruct the frequency table.
8. The scores of 50 students are classified into a box-frequency

distribution table of constant amplitude intervals. Knowing that:y 2 = 50,
n i =4, N2=20, na= 25, c = 62. Rebuild the frame
9. In each of the following cases, determine whether or not the data is

consistent:
a. m = number of classes = 6, h i = 0.2, h4 = 0.2, H2 = 0.6, H3 + H4 =
1.9
b. H 4 = 0.30, n = 10, h 3 = 0.31
c. h 2 =0.40, n = 50, n 1 = 20
dh 1 = 4%,h 3 = 12%,H 4 = 15%
and. H5 = 0.36, N4 = 30, hs = 6, n = 50
10.Suppose that the following distribution table represents the salaries of

civil construction workers in Lima.
to. The civil construction union Daily salaries (in S/.) Frequency
requests that in the new pact Out of 8 to 12 360

collective establishes a of 12 to 16 420
minimum daily wage of S/. 14 of 16 to 20 510
of 20 to 24 660
What percentage of of 24 to 28 570
workers will benefit of 28 to 32 480
with this pact? Total 3000
b. Workers who receive more than 30 soles a day are assumed to be

highly qualified (master builders). That percentage of
workers are supposed to be highly qualified?
c. Estimate the number of workers who earn between 15 and 20

soles a day.

1
0
AVERAGE
The arithmetic mean of a statistical variable is the sum of all its possible
values, weighted by their frequencies.
That is, if the table of values of a variable
x n f
neither fi
xk n.k. fk
The mean is the value that we can write in the following equivalent ways:
X=Xif+ ... + X k f k
= 1 (X n i + ... x k n k )
I
n
1k
= —)X ¡ n ¡
n 1=1
If the data is not arranged in a table, then:
_ X i + ... + X n
x
= n
Observation
We have implicitly assumed in the definition of the mean that we were

dealing with a discrete variable If the variable is continuous we will have
to change the values of X ¡ for the corresponding class marks. In general,
the arithmetic mean obtained from class marks C ¡ , will differ from the
mean obtained from the real values, X ¡ . That is, there will be a loss of
precision that will be greater the greater the difference between the real
values and the class marks, that is, the greater the lengths a i of the
intervals.
Proposition
The sum of the differences of the variable with respect to the mean is
zero, that is:
n
2 (XX)=0
=1
Demonstration:
It is enough to develop the sum to obtain:
n
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X

n )-n x = n x -n x = 0
¡=1
This result tells us that the error made when approximating any value of
the variable, for example X i , using the central value X, is compensated
by the other errors:
n
Approx. error of x i = X1-x=2 (X ¡ - X)
¡=1

1 7
0
If the errors are considered with a positive sign, in this case they cannot
be compensated. This occurs if we take any of the following as error
n
>(X-X)2>0
=1
n
2x - X|20
i=1
max Xi -x| > 0

i=1,.,n
measures:
Quadratic error
Quadratic error
Quadratic error
Which are strictly positive quantities if any
X¡+x
Example IH-I n
1
Obtain the deviations from the mean in the following 0-10 1

distribution and verify that their sum is zero. 10-20 2
Solution 20-30 4
30-40 3
ln - I neither xi xini xi-x (X - x) n i
0 -10 1 5 5 -19 -19

10-20 2 15 30 -9 -18
20-30 4 25 100 +1 +4
30-40 3 35 105 +11 +33
n=
10
2Xn = 240 2=0
The arithmetic mean is
ge1/gn240_24
nL 10
As can be verified by adding the elements in the last
column.
2(Xi -X)n = 0
Proposition (Konig)
For any possible k value that we consider as a
candidate central measure, it improves it in the least
squares sense, i.e.
Introduction............................................................................................2

1 7
0

85+90+ 93+82+95 = 415 = 89 points....................................................9
Issues:...................................................................................................11
Solution:...............................................................................................15
Solution:...............................................................................................16
1. Data collection..............................................................................18
Example:..............................................................................................22
Solution:...............................................................................................23
PRACTICE N° 2..................................................................................25
1.1 Introduction...............................................................................32
Concept:...............................................................................................33

1 7
0

consciousness)......................................................................................40
ll.Quality..............................................................................................40
2.4 Activities to start a process towards total quality...........................41
QA........................................................................................................43
QA........................................................................................................44
allows analysis."...................................................................................45
QA........................................................................................................47
Technical rules.....................................................................................53
ISO 9000 STANDARDS.....................................................................54
ISO 9000 objectives.............................................................................55
ISO 9000 standards..............................................................................55
STANDARD IS014000.......................................................................55

1 7
0
IS014000 Standards.............................................................................56
Solution:...............................................................................................57
Ni=n+n2+na + ... + n i.............................................................................58
Interpretation:.......................................................................................60
Solution................................................................................................66
Observations:.......................................................................................76
Y1-y'i=<y'i-1 , y'i].................................................................77
and. — 93.5494.5 — 94.0.......................................................................79
Solution to:...........................................................................................83
+68......................................................................................................................97
PRACTICE N°3.................................................................................103
Time elapsed from receipt of order to delivery (in days)...................104
AVERAGE.........................................................................................106
Observation........................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0............................................................106
2x - X|20..............................................................................................107
Example.............................................................................................107
Solution..............................................................................................107
ge1/gn240_24............................................................................................107
> 2(xx)2.........................................................................................................114
Observation........................................................................................114
The mean square................................................................................121
THE MEDIAN...................................................................................122

1 7
1
Observation........................................................................................122
Observation........................................................................................122
FASHION..........................................................................................124
Observation........................................................................................125
EXAMPLE.........................................................................................126
= (0,010)= 0,0320.......................................................................145
Breakdown.........................................................................................146
Where:................................................................................................146
Example.............................................................................................146
X=1 2x..............................................................................................................147
a = 12(x-7)2..........................................................................................147
°\4..................................................................................................................147
°s\4............................................................................................................147
PRACTICE N° 4................................................................................148
from $10, from..............................................................................................148
C ¡ xn =ni(ocxh= h) C ¡ Ci................................................149
Solution:.............................................................................................155
Line graphs.........................................................................................158
PRACTICE N° 5..................................................................................159
1. Concept:......................................................................................165
2. Application:................................................................................165
5. Typing.........................................................................................167
_x-u........................................................................................................167
-..............................................................................................................167
— —- z.................................................................................................167
Z
= npq is N(0,1)..........................................................................................168
............................................................................................................168

1 7
1

Example:............................................................................................171
Concept:.............................................................................................173
> 6 2 d...........................................................................................................185
-
138 -:994........................................................................................185
V(a+b)(c+d)(a+c)(b+d).......................................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0........................................................188
5b...........................................................................................................188
2 and xi = b 2x2+ aZx.............................................................188
YsYb(X- x)................................................................................................188
Concept..............................................................................................188
What is it?..........................................................................................189
When it's used?...................................................................................189
How is it used?...................................................................................190
PRACTICE N° 7................................................................................195
Organized:..........................................................................................195
Work stoppage...................................................................................195
Concept:.............................................................................................195
2. Control sheet..................................................................................196

1 7
1
Scatter plots........................................................................................197
Activities:...........................................................................................207
(Conclusions).....................................................................................208
Activities............................................................................................208
n
= £ [(x i -x)- ( and using Newton's binomial...)
(kx)] 2
i=1

1 7
1
APPLIED STATISTICS
n n n
= 2(X - x)2 - 2(x - k) 2(X - x) +2 (k - x) 2
¡=1 ¡=1 ¡=1
EITHER n(kx) > O

2
> 2(xx)2
¡=1
Proposition (Linearity of the mean)

Y = a + bx E>> y = a + bx
Proposition
Given r groups with n i , n2,...nr observations and x i , X2,...Xr being the
respective means of each of them.
Then the mean of the n = n i +... + nr observations is:
_ n i x i + ... + n r x r
x
= n i + ... + n r
Demonstration
Let's call the jth observation of group i Xi; so, we have
XI) /n
1st
group
2nd group
- - - -) xii . .. Xni)
—• ... X2n2 1
K=(Lpi
-
X21
... XI) /n
2
r esmo group - - -• xri ,.. Xm)

.
Thus, conveniently grouping the observations we arrive at: x=( XI) /n
r
(X11 + ... + x i n i ) + (X22 + ... + X2n2) + ... + (x ri + ... + Xrnr)

x =----------------------------------------------------------------------
n i + n 2 + ... + n r
X1+X1+ ... + nrXr
= n
Observation
Despite the good properties that the stocking offers, it has some
drawbacks:
• One of them is that it is very sensitive to extreme values of the

variable: since all observations are involved in the calculation of the
mean, the appearance of an extreme observation will cause the
mean to shift in that direction. Consequently.
• It is not advisable to use the mean as a central measure in highly

skewed distributions;
• It depends on the division into intervals in the case of continuous

variables.
• If we consider a discrete variable, for example, the number of

children in the families of Malaga, the value of the mean may not
belong to the set of values of the variable;
For example x = 25 children.

1
1
Abbreviated calculation
The linearity of the mean can be used to simplify the operations

necessary for its calculation by changing the origin and unit of
measurement. The method consists of the following:
1. We take a number that approximately expresses the type of unit we

work with. For example, if the units we use are millions, we take a =
1,000,000
2. We select any point in the central area of the table, X or . This point
will play the role of reference origin.
3. We change to the variable

__X-X0
X _ XX 0 577 5 to
to E>x = az + X 0
4. In this way we construct the table of the variable Z, for which it is
easier to calculate x directly, and then X is calculated using the
relation (2.2).
Generalized stockings
Depending on the type of problem, various generalizations of the

average
can be considered. Here are some of them applied to observations nails
X i ,...,X n :
The geometric mean
X g is the mean of the logarithms of the values of the variable:

log X 1 + ... + log Xn
log X g = -------n------- —
Then X g = (/X l X 2 ... x n

If the data is grouped in a table, then we have:
The harmonic mean
Xa, is defined as the reciprocal of the 1
arithmetic mean of the reciprocals, that is, —+...+ x
1
For
Xi n
Introduction............................................................................................2
85+90+ 93+82+95 = 415 = 89 points....................................................9
Issues:...................................................................................................11

1 7
1
Solution:...............................................................................................15
Solution:...............................................................................................16
1. Data collection..............................................................................18
Example:..............................................................................................22
Solution:...............................................................................................23
PRACTICE N° 2..................................................................................25
1.1 Introduction...............................................................................32
Concept:...............................................................................................33
consciousness)......................................................................................40
ll.Quality..............................................................................................40

1 7
1
QA........................................................................................................43
QA........................................................................................................44
allows analysis."...................................................................................45
QA........................................................................................................47
Technical rules.....................................................................................53
ISO 9000 STANDARDS.....................................................................54
ISO 9000 objectives.............................................................................55
ISO 9000 standards..............................................................................55
STANDARD IS014000.......................................................................55
IS014000 Standards.............................................................................56
Solution:...............................................................................................57

1 7
1
Ni=n+n2+na + ... + n i.............................................................................58

Interpretation:.......................................................................................60
Solution................................................................................................66
Observations:.......................................................................................76
Y1-y'i=<y'i-1 , y'i].................................................................77
and. — 93.5494.5 — 94.0.......................................................................79
Solution to:...........................................................................................83
+68......................................................................................................................97
PRACTICE N°3.................................................................................103
AVERAGE.........................................................................................106
Observation........................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0............................................................106
2x - X|20..............................................................................................107
Example.............................................................................................107
Solution..............................................................................................107
ge1/gn240_24............................................................................................107
> 2(xx)2.........................................................................................................114
Observation........................................................................................114
The mean square................................................................................121
THE MEDIAN...................................................................................122
Observation........................................................................................122
Observation........................................................................................122
FASHION..........................................................................................124
Observation........................................................................................125
EXAMPLE.........................................................................................126

1 7
1

= (0,010)= 0,0320.......................................................................145
Breakdown.........................................................................................146
Where:................................................................................................146
Example.............................................................................................146
X=1 2x..............................................................................................................147
a = 12(x-7)2..........................................................................................147
°\4..................................................................................................................147
°s\4............................................................................................................147
PRACTICE N° 4................................................................................148
from $10, from..............................................................................................148
C ¡ xn =ni(ocxh= h) C ¡ Ci................................................149
Solution:.............................................................................................155
Line graphs.........................................................................................158
PRACTICE N° 5..................................................................................159
1. Concept:......................................................................................165
2. Application:................................................................................165
5. Typing.........................................................................................167
_x-u........................................................................................................167
-..............................................................................................................167
— —- z.................................................................................................167
Z
= npq is N(0,1)..........................................................................................168
............................................................................................................168
Example:............................................................................................171
Concept:.............................................................................................173

1 7
1
> 6 2 d...........................................................................................................185
-
138 -:994........................................................................................185
V(a+b)(c+d)(a+c)(b+d).......................................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0........................................................188
5b...........................................................................................................188
2 and xi = b 2x2+ aZx.............................................................188
YsYb(X- x)................................................................................................188
Concept..............................................................................................188
What is it?..........................................................................................189
When it's used?...................................................................................189
How is it used?...................................................................................190
PRACTICE N° 7................................................................................195
Organized:..........................................................................................195
Work stoppage...................................................................................195
Concept:.............................................................................................195
2. Control sheet..................................................................................196
Scatter plots........................................................................................197

1 7
2

Activities:...........................................................................................207
(Conclusions).....................................................................................208
Activities............................................................................................208
The mean square

XC, is the square root of the arithmetic mean of the squares.
+ +X2n
/ X21 ...

1 7
2
THE MEDIAN
We consider a discrete variable X whose observations in a statistical

table have been ordered from smallest to largest. We will call median ,
M ed, the first value of the variable that leaves 50% of the observations
below itself. Therefore, if n is the number of observations, the median
will correspond to the observation [n/2] + 1, where we represent the
integer part of a number by [-].
In the case of continuous variables, the classes are given by intervals,

and here the median formula becomes a little more complicated (but not
too much): Let [ln, l ¡ ] be the interval where we have found that 50 are
below % of observations.
The median is then obtained from the accumulated absolute

frequencies, through linear interpolation (Thales Theorem) as follows
(Fig.l).
n -N i-1
CC'_ BB n_2
AC AB 5 a i Med -1«
p-Ni-1
•aini
Med = I ¡- I
+
Observation
The relationship (2.2) corresponds to defining for each possible

observation, x e (l ji , l j ) its accumulated relative frequency, F(x), by linear
interpolation between the values F(l ji ) = F ji and F(l j ) = F j so that
F(x) = F(IH) + F()aFeh) (x - ln)
Thus, Med is the point where F(Med) = 2
This is equivalent to saying that the median divides the histogram into
two parts equal to 2

1 7
2
Observation
Among the properties of the median, we are going to highlight the

following:
• As a descriptive measure, it has the advantage of not being affected by

extreme observations, since it does not depend on the values that the
variable takes, but on their order. Therefore, its use in asymmetric
distributions is appropriate.
• It is quick to calculate and easy to interpret.
• Unlike the mean, the median of a discrete variable is always a value of

the variable we study (e.g. The median of a number of children variable
always takes integer values).
• If a population is made up of 2 subpopulations with medians M edi and

M ed2 , it can only be stated that the median, Med, of the population is
between M edi and M ed2
M edi < M ed < M ed2
• The biggest defect of the median is that it has complicated
mathematical properties, which makes it very difficult to use in
statistical inference.
• It is a function of the chosen intervals.
• It can be calculated even if the lower or upper interval has no limits.
• The sum of the differences of the absolute values of n scores with

respect to their median is less than or equal to any other value. This is
the equivalent of Konig's theorem (proposition 2.í) with respect to the
mean, but where the following is considered as a measure of
dispersion:
n
Med |
>|Xi-
i=yes
Example
Let X be a discrete variable that has presented the modalities on a

sample
X— 2, 5, 7,9,12 -X =7, M ed = 7
If we change the last observation for another abnormally large one, this
does not affect the median, but it does affect the mean:
X->2,5,7,9,í25=^=>-X = 29.6; M ed = 7
In this case the mean is not a possible value of the (discrete) variable,
and has been greatly affected by the extreme observation. This has not
been the case for the median.

1 7
2
Example l ii - l i neither
Obtain the arithmetic mean and the median in the

0-10 60
accompanying distribution. 10-20 80
20-30 30
Graphically determine which of the two averages is 30 - 20
more significant. 100-500 10
Solution:
ln - I neither to i X i xn i Neither neither
0 -10 10- 60 10 5 300 60 60

20 20-30 80 10 15 1.200 140 80
30 10 25 750 170 30
30 -100
20 70 65 1.300 190 2,9
100 -500 10 400 300 3.000 200 0.25
n= ^n i = 6,550
200
The arithmetic mean is
x = 1yx = 6,550 = 32.75

x ,
n— 200
The first accumulated absolute frequency that exceeds the value n/2 =
100 is N=140. Therefore the median interval is [10;20). So
n/2-N ii 100-60
M ed = l i-1 •a i = 10 + x10 = 15
neither 80
+
To see the representativeness of both averages, we make the histogram
in Figure 2 and observe that given the shape of the distribution, the
median is more representative than the mean.
Fig. 2 For this frequency distribution, it is more representative to use the
median as the central tendency statistic rather than the mean.
FASHION
We will call mode any relative maximum of the frequency distribution,

that is, any value of the variable that has a frequency greater than its

1
2
previous and posterior.
In the case of continuous variables it is more correct to talk about modal

intervals. Once this interval, (li-j, li), has been obtained, the following
formula is used to calculate the mode. (Fig. 3).
HC ' _ B'C _ HC + H'C

AB A'B AB+A'B'
F ash - l i-1 _ ai
-
n ¡ - ni- (n i - n ii ) + (n i - n +i )
nn , n i - n ii
=>> Fashion — l ii + /.- .— N,/-------.—{ a i
(n i - n ii ) + (n i - n +i )
Observation
In fashion we highlight the following properties:
• It is very easy to calculate.
• It may not be unique.
• It is a function of the chosen intervals through their amplitude,

number and limits.
• Even if the first or last interval does not have lower or upper
extremes respectively, the mode can be calculated.

1 7
2
EXAMPLE
Problem 1: we have the following information about a frequency

distribution of 50 elements of a material subjected to a breakage test (in
kg/cm 2 ). The length of the class intervals is constant and equal to 20
Iyi-, yi no Neither and not
300
400
23 350
17
- 120 ) 440
50
a. Determine the sample mean and median
b. The n 0 of data that you estimate belongs to the interval [mean,

median]
c. The second decile
Solution: Taking into account that c = 20, and the data in the table, we
complete this
[yi - and' and nei Neith

Yo Yo / and yin
ther er
[20 - 40 ) 30 10 10 300
50 8 18 400
[40-60 )
’ 60 - 80 ) 70 5 23 350
[ 80 - 100 ) 90 17 40 1530
[ 100 - 120 ) 110 4 44 440
[ 120 - 140 ) 130 6 50 780
Totals 50 3800
PARTIAL EVALUATION I

1
2
Introduction............................................................................................2
2. Division of statistics........................................................................8
2.1 Descriptive statistics....................................................................8
85+90+ 93+82+95 = 415 = 89 points....................................................9
Issues:...................................................................................................11
Solution:...............................................................................................15
Solution:...............................................................................................16
1. Data collection..............................................................................18
Example:..............................................................................................22
Solution:...............................................................................................23
PRACTICE N° 2..................................................................................25
1.1 Introduction...............................................................................32
Concept:...............................................................................................33

1 7
2

consciousness)......................................................................................40
ll.Quality...............................................................................................40
QA........................................................................................................43
QA........................................................................................................44
1. "Control is not classification or selection"....................................45
allows analysis."...................................................................................45
QA........................................................................................................47
Technical rules.....................................................................................53
ISO 9000 STANDARDS.....................................................................54
ISO 9000 objectives.............................................................................55

1
2
ISO 9000 standards..............................................................................55

STANDARD IS014000.......................................................................55
IS014000 Standards.............................................................................56
Solution:...............................................................................................57
Ni=n+n2+na + ... + n i.............................................................................58
Interpretation:.......................................................................................60
Solution................................................................................................66
Lower limit Upper limit.....................................................................70
Observations:........................................................................................76
Y1-y'i=<y'i-1 , y'i].................................................................77
and. — 93.5494.5 — 94.0.......................................................................79
Solution to:...........................................................................................83
+68......................................................................................................................97
PRACTICE N°3.................................................................................103
AVERAGE.........................................................................................106
Observation........................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0............................................................106
2x - X|20..............................................................................................107
Example..............................................................................................107
Solution..............................................................................................107
ge1/gn240_24............................................................................................107
> 2(xx)2.........................................................................................................114
Observation........................................................................................114

1 7
2
The mean square................................................................................121

THE MEDIAN...................................................................................122
Observation........................................................................................122
Observation........................................................................................122
FASHION..........................................................................................124
Observation........................................................................................125
EXAMPLE.........................................................................................126
PARTIAL EVALUATION I..............................................................127
RANGE, VARIANCE, STANDARD DEVIATION.........................144
= (0,010)= 0,0320.......................................................................145
Breakdown.........................................................................................146
Where:................................................................................................146
Example..............................................................................................146
X=1 2x..............................................................................................................147
a = 12(x-7)2..........................................................................................147
°\4..................................................................................................................147
°s\4............................................................................................................147
PRACTICE N° 4................................................................................148
from $10, from..............................................................................................148
C ¡ xn =ni(ocxh= h) C ¡ Ci................................................149
Solution:.............................................................................................155
Line graphs.........................................................................................158
PRACTICE N° 5..................................................................................159
1. Concept:......................................................................................165
2. Application:................................................................................165
5. Typing.........................................................................................167
_x-u........................................................................................................167
-..............................................................................................................167
— —- z.................................................................................................167
Z
= npq is N(0,1)...........................................................................................168

1
3

............................................................................................................168
Example:............................................................................................171
Concept:.............................................................................................173
DISPERSION DIAGRAM.................................................................180
> 6 2 d............................................................................................................185
-
138 -:994........................................................................................185
V(a+b)(c+d)(a+c)(b+d).......................................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0........................................................188
5b...........................................................................................................188
2 and xi = b 2x2+ aZx.............................................................188
YsYb(X- x)................................................................................................188
Concept..............................................................................................188
What is it?..........................................................................................189
When it's used?...................................................................................189
How is it used?...................................................................................190
PRACTICE N° 7................................................................................195
Organized:..........................................................................................195
Work stoppage...................................................................................195
Concept:.............................................................................................195
2. Control sheet..................................................................................196

1 7
3
Scatter plots........................................................................................197
A control chart shows.........................................................................198
First Step: Defining the Problem........................................................205
Activities:...........................................................................................207
(Conclusions).....................................................................................208
Activities............................................................................................208
1.
corresponds to an ordinal quantitative variable
2. Quality control in industrial engineering has as its basic component:

SQC, SPC in development and VF administration on a pragmatic
basis.
3. Total quality is a strategy that seeks to guarantee the long-term

survival, growth and profitability of an organization.
Introduction 2
85+90+ 93+82+95 = 415 = 89 points 9
Issues: 11
Solution: 15
Solution: 16
1
3
Example: 22
Solution: 23
PRACTICE N° 2 25
1.1 Introduction 32
Concept: 33
consciousness). 40
ll.Quality 40

1 7
3

QA 43
QA 44
QA 47
General concepts 51
Technical rules 53
Solution: 57
Ni=n+n2+na + ... + n i 58
Interpretation: 60
Solution 66

1
3

Observations: 76
Y1-y'i=<y'i-1 , y'i] 77
and. — 93.5494.5 — 94.0 79
Solution to: 83
+68 97
PRACTICE N°3 103
AVERAGE 106
Observation 106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0 106
2x - X|20 107
Example 107
Solution 107
ge1/gn240_24 107
> 2(xx)2 114
Observation 114
The mean square 121
THE MEDIAN 122
Observation 122
Observation 122
FASHION 124
Observation 125
EXAMPLE 126
= (0,010)= 0,0320 145
Breakdown 146

1 7
3
Where: 146
Example 146
X=1 2x 147
a = 12(x-7)2 147
°\4 147
°s\4 147
PRACTICE N° 4 148
from $10, from 148
Solution: 155
Line graphs 158
PRACTICE N° 5 159
1. Concept: 165
2. Application: 165
5. Typing 167
_x-u 167
- 167
— —- z 167
Z is N(0,1)
= npq 168
168
Example: 171
Concept: 173
>62d 185

1
3

-
138 -:994 185
V(a+b)(c+d)(a+c)(b+d) 186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0 188
5b 188
YsYb(X- x) 188
Concept 188
What is it? 189
When it's used? 189
How is it used? 190
PRACTICE N° 7 195
Organized: 195
Work stoppage 195
Concept: 195
Scatter plots 197

1 7
3
Activities: 207
(Conclusions). 208
Activities 208
4.
financial
5. ISO 9000 standardization consists of formulating and applying rules

with the purpose of putting in order a field of VF activities.
6. The result of a particular or specific VF standardization work is

known as: technical standard
Introduction 2
85+90+ 93+82+95 = 415 = 89 points 9
Issues: 11
Solution: 15
Solution: 16
Example: 22
Solution: 23
PRACTICE N° 2 25
1.1 Introduction 32
Concept: 33

1
3

consciousness). 40
ll.Quality 40
QA 43
QA 44
QA 47

1 7
3

General concepts 51
Technical rules 53
Solution: 57
Ni=n+n2+na + ... + n i 58
Interpretation: 60
Solution 66
Observations: 76
Y1-y'i=<y'i-1 , y'i] 77
and. — 93.5494.5 — 94.0 79
Solution to: 83
+68 97
PRACTICE N°3 103
AVERAGE 106
Observation 106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0 106
1
4
2x - X|20 107
Example 107
Solution 107
ge1/gn240_24 107
> 2(xx)2 114
Observation 114
The mean square 121
THE MEDIAN 122
Observation 122
Observation 122
FASHION 124
Observation 125
EXAMPLE 126
= (0,010)= 0,0320 145
Breakdown 146
Where: 146
Example 146
X=1 2x 147
a = 12(x-7)2 147
°\4 147
°s\4 147
PRACTICE N° 4 148
from $10, from 148
Solution: 155
Line graphs 158

1 7
4
PRACTICE N° 5 159
1. Concept: 165
2. Application: 165
5. Typing 167
_x-u 167
- 167
— —- z 167
Z
= npq is N(0,1) 168
168
Example: 171
Concept: 173
>62d 185
-
138 -:994 185
V(a+b)(c+d)(a+c)(b+d) 186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0 188
5b 188
YsYb(X- x) 188
Concept 188
What is it? 189
When it's used? 189
How is it used? 190

1
4
PRACTICE N° 7 195
Organized: 195
Work stoppage 195
Concept: 195
Scatter plots 197
Activities: 207
(Conclusions). 208
Activities 208
7.
frequencies

1 7
4
RANGE, VARIANCE, STANDARD DEVIATION
MEASURES OF DISPERSION
It studies the distribution of the values of the series, analyzing whether

they are more or less concentrated, or more or less dispersed.
There are various measures of dispersion, among the most used we can
highlight the following:
1. Range: measures the amplitude of the sample values and is

calculated by the difference between the highest value and the lowest
value.
2. Variance: Measures the distance between the values of the series

and the mean. It is calculated as the sum of the squared differences
between
each value and the mean, multiplied by the number of times each
value has been repeated. The sum obtained is divided by the sample
size.
S 2 (X i -X m ) 2
2x
*n¡
The variance will always be greater n than zero. The closer it gets to
zero, the more concentrated the values of the series are around the
mean. On the contrary, the greater the variance, the more dispersed
they are.
3. Standard deviation: It is calculated as the square root of the

variance.
4. Pearson's variation coefficient: it is calculated as the ratio between

the standard deviation and the mean.
Example: we are going to use the data series on the height of the
students in a class and we are going to calculate their dispersion
measures.
Variable Absolute frequencies Relative frequencies

(value) simple Accumulated Simple Accumulated
1,23 1 1 3,3% 3,3%

1,21 4 5 13,3% 16,6%
1,22 4 9 13,3% 30,0%
1,23 2 11 6,6% 36,6%
1,24 1 12 3,3% 40,0%
1,25 2 14 6,6% 46,6%
1,26 3 17 10,0% 56,6%
1,27 3 20 10,0% 66,6%
1,28 4 24 13,3% 80,0%
1,29 3 27 10,0% 90,0%
1,30 3 30 10,0% 100,0%
1. Range: Difference between the highest value in the sample (1.30)

and the lowest value (1.20). Then the range of this sample is 10cm.

1 7
4
((1.20-1.253)2 * 1 + ((1.21-1.253)2 * 4) + ((1.22-

1.253)2*4) + ... + ((1.30-1.253) 2 *3) Sx= 3Ó
Therefore, the variance is 0.0010
3 .- Standard deviation: it is the square root of the variance.
o = (S2.)0a
Then:
= (0,010)= 0,0320
4 . Pearson's coefficient of variation: it is calculated as the ratio
between the standard deviation and the sample mean.
Cv = Ó,Ó32Ó/1.253
Then:
CV = 0.0255
The interest of the coefficient of variation is that since it is a percentage it
allows us to compare the level of dispersion of two samples. This does
not happen with the standard deviation, since it is expressed in the same
units as the series data.
For example, to compare the level of dispersion of a series of data on

the height of students in a class and another series with the weight of
said students, standard deviations cannot be used (one is expressed in
cm. and the other in kg). On the other hand, their coefficients of variation
are both percentages, so they can be compared.
Interpretation and application
The standard deviation is a measure of how dispersed the data is from

the average value. Put another way, the standard deviation is simply the
"average" or expected variation with respect to the arithmetic mean.
A large standard deviation indicates that the points are far from the
mean, and a small deviation indicates that the data are clustered near
the mean.
For example, the three samples (Ó, Ó, 14,14), (Ó, 6, 8,14) and (6, 6, 8,
8) each have a mean of 7. Their standard deviations are 7, 5 and 1,
respectively. The third sample has a much smaller deviation than the
other two because its values are closer to 7.
The standard deviation can be interpreted as a measure of uncertainty.

The standard deviation of a repeated group of measurements gives us
their precision. When determining whether a group of measurements
agree with the theoretical model, the standard deviation of those
measurements is of vital importance: whether the mean of the
measurements is too far from the prediction (with the distance measured
in standard deviations) , then we consider that the measurements
contradict the theory.

1 7
4
APPLIED STATISTICS
This is to be expected since the measurements fall outside the range of

values that would be reasonable to expect to occur if the theoretical
model were correct.
Breakdown
The standard deviation (SD/SD), also known as standard deviation, is

a measure of dispersion used in statistics that tells us how much point
values tend to move away from the average in a distribution. In fact,
specifically the standard deviation is "the average of the distance of each
point from the average." It is usually represented by an S or the letter
sigma, or
The standard deviation of a data set is a measure of how much the data
deviates from its mean. This measure is more stable than the route and
takes into consideration the value of each data.
It is possible to calculate the standard deviation as the square root of the

integral
co
o =J_0 ( x -u) 2 f(x) dx
Where:
CO u=JX f(x) dx
The DS. It is the square root of the variance of the distribution

o 2 = lim 1 2 (xi - x) 2 n-0 nM V
Thus, the variance is the mean of the squares of the differences

between each value of the variable and the arithmetic mean of the
distribution.
Although this formula is correct, in practice it is interesting to make

population inferences, so in the denominator instead of n, n-1 is used
(Bessel Correction)
,Z' (xx)
° n-1
There is also another function that is easier to perform and with less risk
of making mistakes:
Z>2 2
n -1 -x
S2
=
Example
Here's how to calculate the standard deviation of a data set. The data
represents the age of the members of a group of children. {5,6,8,9}

1 7
4
1. Calculate the average

X.
X=1
In this case, N = 4 because there are four
pieces of information:
xi=5
X2 = 6
x3=8
x4=9 4
yx = 1TX
4 •i Replacing N with 4
x = 1 (Xi-X2-X3-Xa)
x = I (5 + 6+8 + 9)
x = 7 This is the average
2. Calculate the standard

deviation
NN ,
G={1
4
2(Xx)2
4
0=1 2 (x i -x) Replacing N with 4
4 i=i
a = 12(x-7)2 Replacing x with 7

N 4 i=i
1 [(4,589 - 4,596) z
+(4,318-4,596) z +(4,256-4,596) z +(4,624-4,596) z +(4,903-4,596) z +(4,867-4,596) z +(4,420-4,596) z +(4.790-
OF\8 43596) z ]
[(5-7)2 + (6 - 7)2 + (8 - 7)2 + (9 -

°\4 7)2]
6={1((-2)2 + (-1)2 + 12+22)
1 (4+1+1+4)
10
°s\4
a =1.5811 This is the standard deviation.

1
4
PRACTICE N° 4
1. The average salary of factory employees is a certain amount. How
would you, as Chief of Staff, respond to a complaint that no one should
earn below the average, whatever it may be?
Answer: The only way to satisfy them is to make everyone earn
equally, which is not possible.
2. Until last semester, a UDEP student had passed 108 of the 141 credits
enrolled, with a cumulative index of 10.43. If you have enrolled in 21
credits this semester, how much can you raise your cumulative index
as a maximum?
Answer: Up to 11.67
3. Can the mean deviation have a value of zero? Can it be negative?
4. What can be said about a distribution in which s = 0?
5. A savings and loan association has the following mortgages with their
respective interest rates: $40,000 at 10%; $25,000 at 9%; $20,000 at
8%; $10,000 at 7% and $5,000 at 6%. What is the average interest
rate paid?
6. In the four classrooms where a Statistics test was taken, there were
14, 27, 27 and 36 students. The means (or averages) in these
classrooms were 14.4; 12.1; 9.9 and 10.2 respectively. What is the
general average?
7. A group of friends who left school together graduated from the UDEP
last semester. Four from Engineering with an average index of 14.95,
six from Business Administration with an average index of 14.12 and
three from Information with an average index of 15.10. What is the
average index of the group? What is the deviation are they giving?
8. Given the following distribution of salaries, in dollars, in a foreign
company:
80 52 92 75 82 96 80
70 90 69 83 94 67 63
61 96 88 63 78 83 99
85 75 81 73 97 109 87
100 85 95 88 98 78 98
98 76 100 58 108 89 84
88 64 81 70 105 64 64
81 91 59 72 97 77 97
a) Build a frequency distribution table, using intervals
from $10, from..........................
b) Determine the mean and standard deviation.
c) Briefly comment on the salary distribution in this company.
d) Construct a cumulative frequency table.
e) Find the salary that 25% of the employees are above.
9. The diameter of 180 screws varies between 0.829 to 1.286 cm.
Suggest a grouping indicating its limits, boundaries and class marks.
10. Consider the following sample: resistance of 50 lots of cotton (Kg.
needed to break a skein)

1 7
4
FREQUENCY HISTOGRAM
It is used to graphically represent the absolute or relative frequency

distributions of continuous quantitative data grouped into classes. The
phenomenon studied is represented by a series of rectangles similar to
those in the bar diagram; However, the histogram bars are placed only
vertically and must be next to each other without a space separating
them. The base of each rectangle is the class width of the corresponding
variable. These graphs are called absolute frequency histograms (or
relative frequency histograms). They are constructed as follows:
1. The limits of the class intervals y'o, y',...y'm are taken on a horizontal
axis
2. Above each class interval, rectangles are erected whose area is

exactly the frequency absolute (or relative)
correspondent.
That is, to the class interval [y'-' y' >
with amplitude Ci, and absolute
frequency ni (or relative frequency hi),
a rectangle with base C ¡ and height
n/c (or respectively hi/ci) will
correspond. , which guarantees that
the area of said rectangle, that is, base
times height, coincides with ni (or hi as
the case may be). Fig. Yo
C ¡ xn =ni(ocxh= h) C ¡
Ci
Thus, the total surface area of the y' i-1 AN
rectangles or e! area limited by e! D
histogram will be equal to an (or unity), C i
which allows two different distributions of
effects to be visually compared.
Note: When the class intervals are of equal amplitude as has been
considered in the examples (ci = c), the heights of the ni/c rectangles
(oh/c) will be proportional to the absolute (or relative) frequencies.
In this case it is customary to take the heights numerically equal to said

frequencies (that is, n/c is changed to noh/c to h¡). This is equivalent to
putting c one unit.
Example 1: Figures 2 and 3 would be the histograms of absolute and
relative frequencies respectively for the frequency distribution of the
weights of the steel ingots of example i, Figure 3 .
Note that the only difference between both histograms is the vertical
scale. The scale of Fig. 2 is the absolute number of observations in each
class; On the other hand, the scale of Fig. 3 is the number of
observations in each class as a fraction of the total number of them. That
is, both graphs have the same shape.

1 7
4
Fig. 2 Absolute Frequency Histograms of Steel Ingot Weights
Fig. 3 Relative Frequency Histogram of the weights of steel ingots.
It is useful to be able to present the data in terms of the relative frequency

of observations and not in terms of the absolute frequency of each class,
since, although the absolute numbers may change (for example, by
weighing more steel ingots), the relationship between classes can remain
stable. 40% of ingots can fall into the "9.35 - 94.5" class, regardless of
whether we weigh 50 or 500 ingots.
Usefulness of Graphs: Graphs of absolute frequency distribution and

relative frequency distribution are useful because they highlight and
clarify trends that are not easily captured in tables. They draw the
reader's attention to trends in the data. They allow us to estimate some
values at a glance and provide us with a graphical verification of the
veracity of our solutions.

1 7
5
Frequency polygons
The polygons of absolute or relative frequencies are obtained by joining

the midpoints of the upper bases of the rectangles in the histogram of
absolute or relative frequencies, respectively. Thus, figures 4 and 5 would
be the absolute and relative frequency polygons for the distribution of the
weights of steel ingots in example 1.
Fig. 4. Polygon of absolute frequencies of the weights of 50 steel ingots
If we compare these figures with Figures 2 and 3, we see that a class

interval of zero observations has been added to each end of the
distribution, allowing the polygon to reach the horizontal axis at both
ends.
Fig. 5 Relative frequency polygon of the weights of 50 steel ingots.
It could be shown that the area enclosed by the rectangles of the

histogram coincides with that enclosed by the frequency polygon.
Therefore there would be no difference between both graphs.
However, when you want to present more than one statistical series in the
same graph for comparative purposes, frequency polygons should be
used.

1 7
5
Warhead cumulative frequency polygons
An ogive is the graphical representation of a distribution of cumulative

absolute frequencies or cumulative relative frequencies.
OGIVE "LESS THAN": It is the graphic representation of a distribution of

accumulated absolute frequencies "less than" or cumulative relative
frequencies "less than". They are constructed as follows:
1. The limits of the class intervals y o , y i ,_y m are carried on a horizontal

axis
2. A segment is raised above the upper end of each class whose length
coincides with the accumulated absolute frequency or with the
accumulated relative frequency of said class, depending on whether
we are considering an ogive of accumulated absolute frequencies or
an ogive of accumulated relative frequencies.
3. Then the upper ends of said vertical segments are joined with straight
line segments.
Example 2: Construct the "less than" ogive of cumulative absolute

frequencies and cumulative relative frequencies for the distribution of
steel ingot weights in Example 1.
Solution: We write the distribution tables of cumulative absolute
frequencies and cumulative relative frequencies as follows:
Accumulated Cumulative
Weights (in Kg) class
absolute relative
interval
frequency Ni frequency Hi
Menorque 91.5 either 0.00
Menorque 92.5 4 0.08
Menorque 93.5 15 0.30
Menorque 94.5 35 0.70
Menorque 95.5 44 0.88
Menorque 96.5 50 1.00
Fig. 6 Ogive “Less than” weight distribution of steel ingots using cumulative absolute frequencies
Then, following the given

guidelines, the “less
than” warheads are built,
Figures 6 and 7.
Note: Percentage warhead
If, in an ogive of accumulated relative frequencies, the cumulative relative

frequency scale is changed on the vertical axis to the percentage scale,
the percentage ogive is obtained.
Note that the "less than" ogive is always crescent or horizontal.
OGIVE "GREATER OR EQUAL THAN" ; It is the graphic representation

of a distribution of cumulative absolute frequencies; "greater than or equal
to" or the cumulative relative frequencies "greater than or equal to." It is
constructed completely similar to the lesser than warhead.

1 7
5
Fig. 7 “Less than” ogive of steel ingot weight distribution using cumulative relative frequencies
Example 3: Construct the largest ogive as, from the cumulative absolute
frequencies and cumulative relative frequencies for the steel ingot weight
distribution of Example 1.
Solution: frequency distribution tables greater than or equal to what is

written:
Weight (in Kg) class -* *Hello
interval Neither
Greater than or equal to

Then, following the same process
50 1.00
91.5 46 0.92 of the warheads less than,
Greater
Greater than
than 92.5
93.5 35 0.70 They build the warheads greater
Greater than 94.5 15 0.30 that. (Figs.8 and 9)
Greater equal to 95.5 6 0.12
Greater than 96.5 0 0.00
Note that the "greater than or equal to" ogive is always decreasing or
horizontal.
The main advantage of the ogive is the ease with which it can be linearly
interpolated between two plotted points. For example, the researcher
might be interested in the percentage of steel ingots that weigh less than
a specific value, say 95.1 kg (i.e., given the abscissa, find the ordinate).

1 7
5
Nr
50
45
40
35
30
25
20
15
10
5
Fig. 8 Ogive “greater than or equal to” steel ingot weight distribution, using cumulative
absolute frequencies.
Fig. 9 Ogive “greater than or equal to” steel ingot weight distribution, using cumulative relative
frequencies.
A vertical line that intersects the "less than" curve is drawn along the
known abscissa and a horizontal line is drawn through this point. The
approximation to the desired percentage is obtained by reading the point
of intersection of the horizontal with the axis of the vertical scale. Fig. 10.
In this case, approximately 80% of the ingots will weigh less than 95.1 Kg.
This implies that 20% of the ingots weigh 95.1 Kg or more. This is
obtained by drawing a vertical line for 95.1 Kg until it intersects the
"greater than or equal to" curve, then drawing a horizontal and reading
the point of intersection of this horizontal with the axis of the vertical scale
Fig. 10. (Note the direction of the arrow.)
If, on the other hand, the researcher wants to know below what value a
certain specific percentage of the distribution of observations is located
(that is, known ordinate, find the abscissa). For example, if you want to
know what value 25% of the weight of steel ingots is below.
PRODUCTION MECHANICS 9/113 99/147

15
4
Fig. 10 Graph of both warheads the “less than” and the “greater than or equal to”
Through this point a horizontal is drawn that intersects the "less than"
curve and a vertical is drawn through the intersection. The approximation
to the desired value is obtained by reading the point of intersection of the
vertical with the abscissa axis (Fig. 10). In this case it is approximately
93.3 Kg. That is, 25% of steel ingots weigh less than 93.3 Kg.
An important conclusion that can be drawn from the graph of both

warheads (Fig. 10) is that these intersect at a point "M" that corresponds
to 50% of the observations. For our example this point is approximately
94 Kg.
Example 4: The economic magazine "Peru in figures" has collected the
following data from a sample of 130 sellers that represent the total annual
income from commission:
<Utilities (in S)] Frequency a. Construct a frequency histogram with this
5,000 or less 5 data.
5 000- 10 000 9 b. What do you notice in the histogram
10 000- 15 000 11
15 000-20 000 33 about the data that you didn't
20 000 - 30 000 37 immediately notice in the frequency
30 000 - 40 000 19
40 000 - 50 000 9 distribution?
more than 50,000 7 c. Build an ogive that helps you answer the
following questions:
I. Approximately what proportion of sellers earn more than $25,000?
II. Approximately how much does the "average" salesperson in the

sample earn?
III. Approximately how much could a real estate agent earning 25% of the
maximum earn per year?
Solution:
a. The class intervals have different amplitudes. So to plot the histogram,

we first standardize the amplitude, taking c = 5,000 = unit as the unit.
The calculations can be seen in the following table.

1 7
5
Height = n
< profits (in $)] Frequency n Class Breadth Standardized
i
c
amplitude i/
c = 5000 = unit
c¡ standar
5,000 or less 5 — — —
5 000- 10 000 9 5 000 1 9
10 000- 15 000 11 5 000 1 11
15 000-20 000 33 5 000 1 33
20 000 - 30 000 37 10 000 2 18.5
30 000 - 40 000 19 10 000 2 9.5
40 000 - 50 000 9 10 000 2 4.5
more than 50,000 7 — — —
Note that the extremes are open intervals. Its extent is not known;
Therefore, these intervals cannot be standardized nor graphed.
seven sellers
winning
50 thousand or
more
35 40 45 50 income.
(in thousands of $)
Fig. 11 Absolute frequency histogram of the profits of 130 salespeople
b. The lower end of the distribution is flatter than the upper end.
c. We will construct the “less than” cumulative relative frequency ogive.
Relative frequency Cumulative relative

Income (in $) neither
(
hi) frequency Hi
< 5 000 5 0.038 0.038

5 000- 10 000 9 0.069 0.107
10 000- 15 000 11 0.085 0.192
15 000-20 000 33 0.254 0.446
20 000 - 30 000 37 0.285 0.731
30 000 - 40 000 19 0.146 0.877
40 000 - 50 000 9 0.069 0.946
5 50 000 7 0.054 1.000
TOTAL 130 1.000
Yo. Through point 25 a vertical is drawn that cuts the ogive and a
horizontal through this. The value is approximately 42%. That is,
approximately 42% of sellers earn more than $25,000.

1 7
5
This solution, as we know, can also be done by the linear interpolation

method. In this method it is assumed that the frequencies within every
interval are uniformly distributed. And absolute or relative frequencies
can be used indifferently.
1. using absolute frequencies.
x30
20 40 50
37 19
„ ' . (30-25)37 .............................

30 - 25 — x, Then x =---------------10- =5x3.7= 18.5 = 19
Therefore: N = 19 + 19 + 9 + 7 = 54
N54
The percentage will be: P =--x 100 =---------x 100 « 42%
130 130
2. Using relative frequencies:

25
20 30 40 50
28.5% 14.6% 6.9% 5.4%

1 7
5
APPLIED STATISTICS
(30 - 25) 28.5%

P = 1-------10--------+ 14.6% + 6.9% + 5.4%
= 14.25% + 14.6% + 6.9% + 5.4% = 41.15%
The difference is due to the rounding
ii. The average salesperson will receive 50% of the frequency.
The approximate value that corresponds is $22,000
iii. The 25% of the upper level implies the lower 75%, the current one
corresponds to approximately $31,000
Line graphs
Line graphs are those that are represented on Cartesian coordinate axes
by means of a straight or broken line, since the values of the statistical
variables will be isolated, both in the case of discrete variables and in that
of continuous variables, because the domain is divided into classes and a
class brand is chosen.
Frequency polygons and ogives are examples of line graphs. These

graphs are also used to represent chronological series, that is,
distributions analyzed over time.
In chronological series, time (years, months, days, etc.) is located on the

horizontal axis, called the chronological scale. On the vertical axis, called
the numerical or value scale, the value of the data is represented. Both
scales are independent.
The graph can represent a single variable as a function of time or more

than one variable as a function of time.
Example 5: The number of pieces (in thousands) existing in a Company's
warehouse on the last day of each month of the year 1990 is given in the
following table. Represent a line graph with this data.

15 7
8
Note: The graph of a symmetric distribution can be folded along a vertical

axis so that the two sides coincide. A distribution whose graph does not
have this characteristic is known as skewed. It is said that the distribution
illustrated in Fig. 14 is skewed to the right, because it has a long tail on
the right side and a much shorter tail on the left side. In Fig. 15 it is
observed that the distribution is symmetrical, while Fig. 16 is skewed to
the left.
Fig. 14 Distribution skewed to the right
PRACTICE N° 5
1. Establish the truth or falsehood of each of the following propositions.

Replace each false statement with the corresponding true statement.
a. The vertical scale of an ogive for a relative frequency distribution

marks the fraction of the total number of observations that fall within

1 7
5
each class.
b. The histogram is a series of rectangles, each width proportional to

the number of elements that fall within a particular class of data.
c. The "greater than or equal to" ogive has the S shape and a
negative slope.
d. An advantage of the histogram over a frequency polygon is that it

more clearly shows each individual class in the distribution.
2. Which of the following statements about histogram rectangles is

correct?
a. The rectangles are of height proportional to the number of elements

that fall within the classes.
b. There are generally 5 rectangles in each histogram.
c. The area of a rectangle depends on the number of elements in the
class compared to the number in the rest of the classes.
d. All of the above.
e. aycperonob.
3. In the office of the newspaper El Peruano, the time it takes to print the
front page was recorded for 50 days. The data is transcribed below,
approximated to tenths of a minute:
20. 20.8 22.8 21.9 22. 20. 20. 25. 22. 22.
19. 25.3 20.7 22. 21. 23. 23. 20. 22. 23.
21. 23. 20.3 23. 19. 25. 25. 19. 24. 24.
23. 21. 21.5 23. 19. 24. 24. 19. 23. 22.
22. 19. 24.2 23. 20. 23. 24. 21. 20. 21.
a. Construct a frequency distribution and a “less than” cumulative
frequency distribution from the data, using 0.8-minute intervals.
b. Build a frequency polygon with the data.
c. Construct a "less than" frequency warhead from the data.
d. Using the ogive, estimate what percentage of the time the front page
of the newspaper can be printed in less than 24 minutes.

1 7
6
4. The following distribution of the ages of people hospitalized in a

surgery room. Draw the histogram and the corresponding polygon.
5 . For the following distribution Years Mortality (%)

of infant mortality rates, draw its
0 - 0.20 50.0
histogram and frequency 0.20 - 0.50 30.0
polygon. 0.50 - 1.00 12.0
1.00-2.00 05.0
2.00 - 3.00 03.0
Interval (age in years) Frequency
27-32 2
32-37 5
37-47 6
47-62 3
Total 16
6. The following histogram was the result of a study carried out at an

automotive mechanics school on the repair time of an engine.
a. How many repairs are included in this study?
b. How many repairs were completed in a time greater than 11.5 hours?
What percentage of the repairs were made in a time greater than 11.5
hours?
c. What percentage of the repairs were completed in a time period

between 7.5 and 11.5 hours?
7. The histogram shown below represents the heights (in centimeters) of

the children who are members of the athletics team 8 of an elementary
school.

1 7
6
a. Reconstruct the frequency

distribution table.
b. How many children are on the

team?
c. Children height 145 or taller make

up the basketball team. How many
are in this team?
8. The IQs of a group of 8 students

have been measured using
appropriate tests, with the results grouped into 6 intervals of variable
amplitude. If these amplitudes are: c1 = 4, c2 = 12, c3 = 4, c4 = 12, C5
= 8, c6 = 20. If the cumulative relative frequencies are: H1 = 0.20, H2 =
0.20, H3 = 0.50, H4 = 0.70, H5 = 0.85.
a. Build the distribution table of absolute frequencies (relative,

cumulative absolute and cumulative relative), if the lower limit of the
first class is 80.
b. Draw a Frequency Histogram.
c. Draw a "greater than or equal to" and "less than" ogive.
d. Determine the point of intersection of the warheads.
9. From a children's hospital there is the following report of care per day:
up to 2 months of age, 40 children; from 2 to 6 months, 60 children,
from 6 to 1 year, 60 children; from 1 to 2 years, 96 children and from 2
to 3 years, 72 children. Represent this distribution on a frequency
histogram and draw the frequency polygon.
10. Prepare a bar graph showing the six largest American countries:
Brazil, 8,511,965 km 2 , Canada, 9,976,137 km 2 , United States,
9,363,498 km 2 , Argentina, 2776,889 km 2 , Mexico, 1,958,201 km 2 ,
Peru , 1285215 Km 2 .
11. Show in a bar chart the sales and costs of a warehouse that for the
last three years has the following figures reported by the Accounting
Department:
Warehouse costs of
Year Sale
merchandise sold
1988 S/. 58 6474 36 3252

1989 47 6892 33 5817
1990 59 9886 37 2338

1 7
6
12. Show in a pictogram the inhabitants (in thousands) of five cities

(A, B, C, D and E) which in 1988 TO 350 D400
b 500 AND
c 600
13. Represent the data in the following table using multiple bar diagrams:
PERU: ECONOMIC CRIMES RECORDED BY THE POLICE
OF INVESTIGATIONS: 1986 - 1988
ECONOMIC CRIMES
TOTAL
ADULT SPECULATE
HOARD I LIE OTHERS
RATION CION
1986 1945 230 614 457 644

1987 3616 236 480 1290 1610
1988 13582 2708 3149 4025 3700
Source: Ministry of the Interior - Statistical Directorate

14. Represent the data in the following table using a multiple bar
diagram:
PERU: SERVICE PROVISIONS OF THE PERUVIAN INSTITUTE
SOCIAL SECURITY BY TYPE OF ACTIVITY: 1987 - 89
(IN THOUSANDS)
TYPE OF ACTIVITY
YEAR MEDICAL DENTAL
CARE VACCINES BIRTH HOSPITALS
CARE.
(CONSULTATI (DOSE) (BIRTH) EGRESS DAYS
(SESSION)
ON)
1987 7110 950 278 46 183 2134

1988 8641 1253 330 52 212 1956
1989 8154 1373 518 59 197 1675
Source: PERUVIAN INSTITUTE OF SOCIAL SECURITY - Statistical Directorate
15. Represent the production of rice, cotton and the total production in
the following table using a composite or proportional bar diagram.
PERU: CROPPED SURFACE WITH LOANS APPROVED BY THE AGRICULTURAL BANK WITH
OWN RESOURCES: 1970 - 88 (Has)
AVAILABLE SURFACE WITH LOAN FROM THE AGRARIAN BANK
Years Sugar Sorghum
Total Cotton Rice Coffee Bean Corn 1/ Potato Wheat Others
cane 2/
1980 528707 139524 130156 49939 8183 13549 68591 38776 21167 4300 54522
1981 565516 119133 162598 35577 11484 18357 84889 50053 15317 3364 64726
1982 548337 105138 159291 36883 35002 13569 79303 38899 15347 1787 63118
1983 462149 74478 165060 37276 5104 4929 61965 46024 13266 805 53242
1984 600790 111302 197488 40888 7178 11472 82738 59494 15562 1512 73156
1985 681457 184041 182277 49328 9700 10442 78653 61986 4477 9340 91213
19861039798 109388 209038 92202 18348 24904 212831 152766 7052 23557 189712
19871056549 108010 217726 62216 5549 16048 220529 166417 8144 22171 229739
1988 945752 134145 174438 70669 4908 13178 185559 131798 10827 22047 198183
Note: In 1988, the Third Party Account is considered.

1/ CORN: Yellow, starchy, chala and corn.
2/ SORGHUM: Grain and broom.
SOURCE: BANCO AGRARIO DEL PERU - Office of Planning and Economic Studies.

1 7
6
APPLIED STATISTICS
16. Represent the data in the following report using a pie chart. In an
industry the accounting report shows that production was S/. 62600
and the following expenses: administration S/.11160, materials and
energy S/. 15650, salaries and
benefits S/.18780.
17. Represent the data in the following table using a pie chart for each
year.
FOB EXPORTS BY PRODUCT GROUP (Millions of US dollars)
1984 1985 1986
I Traditional Products
Miners 1368 1205 1023
Oil and byproducts 618 645 236
Agricultural 198 225 333
Fishing boats 137 118 204
Other traditional 100 71 65
II Non-Traditional Products 726 714 648
Total 3147 2978 2509
Source: Central Reserve Bank of Peru
18. The profits of the company ROS, SA (in thousands of soles) have
been:
1980 200 1986 425
1981 250 1987 400
1982 250 1988 400
1983 300 1989 300
1984 350 1990 350
1985 400 1991 400
represent this data in a line graph.
19. If you. He is a publicist and has to graphically present the following
data relating to three preparation academies in the Admission Exam:
ACADEMY N ENTRANTS
0
NONO ENTRANTS
TO 300 1200
b 600 1800
c 100 400
a. How would you graph "academy size and Admissions result"?
b. Which graphical representation is most appropriate for comparing the
"degree of success" of academies?
20. In an industrial company, the payroll by class in a certain month was
like this: administration personnel, $423,380; sales staff, $560,325;
production staff, $2360420. Create a chart that highlights:
1 . The percentage of each estate based on the total payroll,
2 .The percentage of the payroll of each class and the total based on
total sales, which was $12885518.
NORMAL DISTRIBUTION

16 7
4
1. Concept:
This distribution is frequently used in statistical applications. Its name

itself indicates its widespread use, justified by the frequency or
normality with which certain phenomena tend to resemble this
distribution in their behavior.
Many continuous random variables have a density function whose

graph is bell-shaped.
2. Application:
The importance of the normal distribution is mainly due to the fact that
there are many variables associated with natural phenomena that
follow the normal model.
• Morphological characters of individuals (people, animals, plants,...) of

a species, p. e.g. Sizes, weights, wingspans, diameters, perimeters...
• Physiological characteristics, for example; effect of the same dose of

a drug, or the same amount of fertilizer.
• Sociological characteristics, for example: consumption of a certain

product by the same group of individuals, test scores.
• Psychological characteristics, for example: IQ, degree of adaptation

to an environment...........
• Errors made when measuring certain magnitudes.
• Master statistical values, for example: the mean.
• Other distributions such as the binomial or the Poisson distribution

are normal approximations...
And in general any characteristic that is obtained as a sum of many

factors.

1 7
6
3. Density Function
Using rather laborious calculations, it can be shown that the model of

the density function corresponding to such distributions is given by the
formula.
Density Function Graphical representation of this density function
fr- H) 2
1 2 or
2
H = mean or
= dev. typical
1=3.1415 e
= 2.7182... x
or 2 = = abscissa
variance
Domain: Domf=lR
Maximu (or,-)
m: or ^j2
Q. Inflection: enx= u+ or
yx=uG Asymptotes:
The OX axis is a horizontal
asymptote
Symmetries: With respect to the line x =
u
Monotony: Increasing (-oo, u), decreasing
2
Q.
Court:
The normal distribution is defined by two parameters, its mean and its
standard deviation, and we represent it like this
N(u,G)
For each value of u and o we will have a different density function
Therefore the expression N (u,0) represents a family of distributions
normal
4. Distribution function
• It can take any value (-, +)

• Values close to the central one, which we call the mean, are more
likely.
• As we we separate of that worth, probability goes
decreasing equally to the right and left (it is symmetrical).
• As we we separate of that worth, probability goes
decreasing from way more or less fast according to a
parameter, which is the standard deviation.
F(x) is the shaded area of this graph

Distribution Function
(xu ) . - -- 2
2
x1 2 or ,
^( x ) = e—e dx
-^oy/2 T
F(x) = P(X < x)

1 7
6
5. Typing
If the variable X is N(u,0) then the typed variable of
_x-u........................ ....................
Z =—-— And it also follows a normal distribution but
Deu=Oy O = 1,i.e. N(0, 1)
Therefore its density function is:
z2
-
1 2
9 (z) = 2,* e ;- 0 <z< + 0
And its distribution function is

t2
— —-
z
p Z
F(z) = ( < z) = %(2) =2. i dt e
Being the graphical representation of this function.
The variable Z is called the standard variable of X, and the curve of its
density function is called the standard normal curve.
Characteristic of the standard (reduced, standard) normal

distribution
* Does not depend on any parameter

• Its mean is 0, its variance is 1 and its standard deviation is 1.
* The f(x) curve is symmetrical with respect to the OY axis
• Has a maximum on this axis
• It has two inflection points at z=1 and z = -1
Approximation of the Binomial by the Normal ( De Moivre 's
Theorem ):
He showed that under certain conditions (for large n and both p and q are
not close to zero) the Binomial distribution B(n, p) can be approximated
by a normal distribution.
_X-np N (np,npq), De Moivre's

Z
= npq is N(0,1) Theorem
and, therefore, the variable

1 7
6
APPLIED STATISTICS
We must take into account that the greater the value of n, and the closer
it is to 0.5, the better the approximation made will be. That is, it is
enough to verify it.
np > 5ynq > 5
Thanks to this approximation it is easy to find binomial probabilities,
which for large values of n are very laborious to calculate.
It must be taken into account that to correctly carry out this

transformation of a discrete (binomial) variable into a continuous
(normal) variable, it is necessary to make a continuity correction.
P(X = a) = P(a-0.5 < X < a + 0.5)
P(X<a) = P(X < a-0.5) P(X < a) = P(X < a + 0.5)
t
o
P(a < X < b) = P(a-0.5 < X < b + 0.5) P(a <
ab
Theorem of
DE MOIVRE
X is B(n, p) is N(np,npq)
6. Table management, most frequent cases
The distribution of the variable Z is tabulated

16 7
8

1 7
6
P(-a<Z<b) =P (Z < b)-P(Z < a)
P (-bZ >a) =1- (Z < a)
Example:
Let us consider that the weight of Spanish male children at the time of
birth is normally distributed. If we know that the average weight at birth is
3.25 kg and the standard deviation is 0.82 kg, what is the probability that
the weight of a male child at birth is greater than 4 kg?
We typify the random variable x, children's birth weight. In the typing

process, the value of x=4 corresponds to the value, t=0.9146 :
In the table of the standard normal distribution, we look for the value of a
corresponding to the value of t=0.9146 ; The probability of t > 0.9146 is,
as can be seen in the figure: a /2. Then:
p(x>4) = p(t> 0.9146) =036 = 0.18
The standard curve

1 7
7
+ 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.08 0.08 0.09
0.00 .5000 .4960 .4920 .4880 .4840 .4801 .4761 .4721 .4681 .4641
0.10 .4602 .4562 .4522 .4483 .4443 .4404 .4364 .4325 .4286 .4247
0.20 .4207 .4168 .4129 .4090 .4052 .4013 .3974 .3936 .3897 .3859
0.30 .3821 .3783 .3745 .3707 .3669 .3632 .3594 .3557 .3520 .3483
0.40 .3446 .3409 .3372 .3336 .3300 .3264 .3228 .3192 .3156 .3121
0.50 .3085 .3050 .3015 .2981 .2946 .2912 .2877 .2843 .2810 .2776
0.60 .2743 .2709 .2676 .2643 .2611 .2578 .2546 .2514 .2483 .2451
0.70 .2420 .2389 .2358 .2327 .2296 .2266 .2236 .2206 .2177 .2148
0.80 .2119 .2090 .2061 .2033 .2005 .1977 .1949 .1922 .1894 .1867
0.90 .1841 .1814 .1788 .1762 .1736 .1711 .1685 .1660 .1635 .1611
1.00 .1587 .1562 .1539 .1515 .1492 .1469 .1446 .1423 .1401 .1379
1.10 .1357 .1335 .1314 .1292 .1271 .1251 .1230 .1210 .1190 .1170
1.20 .1151 .1131 .1112 .1093 .1075 .1056 .1038 .1020 .1003 .0985
1.30 .0968 .0951 .0934 .0918 .0901 .0885 .0869 .0853 .0838 .0823
1.40 .0808 .0793 .0778 .0764 .0749 .0735 .0721 .0708 .0694 .0681
1.50 .0668 .0655 .0643 .0630 .0618 .0606 .0594 .0582 .0571 .0559
1.60 .0548 .0537 .0526 .0516 .0505 .0495 .0485 .0475 .0465 .0455
1.70 .0446 .0436 .0427 .0418 .0409 .0401 .0392 .0384 .0375 .0367
1.80 .0359 .0351 .0344 .0336 .0329 .0322 .0314 .0307 .0301 .0294
1.90 .0287 .0281 .0274 .0268 .0262 .0256 .0250 .0244 .0239 .0233
2.00 .0228 .0222 .0217 .0212 .0207 .0202 .0197 .0192 .0188 .0183
2.10 .0179 .0174 .0170 .0166 .0162 .0158 .0154 .0150 .0146 .0143
2.20 .0139 .0136 .0132 .0129 .0125 .0122 .0119 .0116 .0113 .0110
2.30 .0107 .0104 .0102 .0099 .0096 .0094 .0091 .0089 .0087 .0084
2.40 .0082 .0080 .0078 .0075 .0073 .0071 .0069 .0068 .0066 .0064
2.50 .0062 .0060 .0059 .0057 .0055 .0054 .0052 .0051 .0049 .0048
2.60 .0047 .0045 .0044 .0043 .0041 .0040 .0039 .0038 .0037 .0036
2.70 .0035 .0034 .0033 .0032 .0031 .0030 .0029 .0028 .0027 .0026
2.80 .0026 .0025 .0024 .0023 .0023 .0022 .0021 .0021 .0020 .0019
2.90 .0019 .0018 .0018 .0017 .0016 .0016 .0015 .0015 .0014 .0014
3.00 .0013 .0013 .0013 .0012 .0012 .0011 .0011 .0011 .0010 .0010
3.10 .0010 .0009 .0009 .0009 .0008 .0008 .0008 .0008 .0007 .0007
3.20 .0007 .0007 .0006 .0006 .0006 .0006 .0006 .0005 .0005 .0005
3.30 .0005 .0005 .0005 .0004 .0004 .0004 .0004 .0004 .0004 .0003
3.40 .0003 .0003 .0003 .0003 .0003 .0003 .0003 .0003 .0003 .0002
3.50 .0002 .0002 .0002 .0002 .0002 .0002 .0002 .0002 .0002 .0002
3.60 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001
3.70 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001
3.80 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001
3.90 .0000 .0000 .0000 .0000 .0000 .0000 .0000 .0000 .0000 .0000
The table above shows the calculated values for:

+ go
2.
t dz=A
The previous table was

calculated using the following expression
A = mo(tobs - tuff + tuff - tob2 + t o+ b i )
Where:
2
m o = 0.39894228 * 2.71828183-4 to =
2 1+p or z
P o = 0.23164190 b2 = 0.35656378 b4 = 1.82125598

b i = 0.31938153 b3 = 1.78147794 b5 = 1.33027443
(this applies to non-negative Zs)
As an example we will calculate the area under the curve from 0.98 to
infinity.

1 7
7
APPLIED STATISTICS
CAUSE-EFFECT DIAGRAM
Concept:
The cause-effect diagram is a way to organize and represent the

different theories proposed about the causes of a problem. It is also
known as Ishikawa diagram or fishbone diagram and is used in the
Diagnosis and Solution phases of the cause.
How to interpret a cause-effect diagram:
The cause-effect diagram is a vehicle to organize, in a very concentrated

way, all the causes that can supposedly contribute to a certain effect. It
allows us, therefore, to achieve common knowledge of a complex
problem, without ever being a substitute for data. It is important to be
aware that cause-effect diagrams present and organize theories. Only
when these theories are tested with data can we prove the causes of
observable phenomena.
Common errors are constructing the diagram before analyzing the

symptoms globally, limiting the proposed theories by unintentionally
masking the root cause, or making errors both in the causal relationship
and in the order of the theories, involving a significant expenditure of
time.
Examples of cause-effect
Below we will see how the value of a quality characteristic depends on a

combination of variables and factors that condition the production
process (among other processes). We are going to use the example of
making mayonnaise to explain Cause-Effect Diagrams:
The variability of quality characteristics is an observed effect that has

multiple causes. When a problem occurs with the quality of the product,
we must investigate to identify the causes of the problem. To make a
Cause-Effect Diagram we follow these steps:

17 7
2
The seven tools of quality

H1 - Cause-Effect Diagrams
We have seen in the introduction how the value of a quality characteristic

depends on a combination of variables and factors that condition the
production process. Let's continue with the mayonnaise manufacturing
example to explain the Cause-Effect Diagrams:
The variability of quality characteristics is an observed effect that has

multiple causes. When a problem occurs with the quality of the product,
we must investigate to identify the causes of the problem.
For this we use Cause-Effect Diagrams, also known as Fishbone

Diagrams due to the shape they have. These diagrams were first used
by Kaoru Ishikawa.
To make a Cause-Effect Diagram we follow these steps:
1. We decide which quality characteristic we are going to analyze will

be. For example, in the case of mayonnaise it could be the weight of
the full jar, the density of the product, the percentage of oil, etc.
We draw a thick arrow that represents the process and to the right we
write the quality characteristic:
Characteristic
quality
2. We indicate the most important and general causal factors that can
generate the fluctuation of the quality characteristic, drawing
secondary arrows towards the main one. For example, Raw Materials,
Equipment, Operators, Measurement Method, etc.

1 7
7
3. We incorporate in each branch more detailed factors that can be

considered causes of fluctuation. To do this, we can ask ourselves
these questions:
a. Why is there fluctuation or dispersion in quality characteristic

values? Due to the fluctuation of Raw Materials. Raw Materials are
noted as one of the main branches.
b. Which Raw Materials produce fluctuation or dispersion in the

values of the quality characteristic? Oil, Eggs, salt, other
seasonings. Oil is added as a minor branch of the main Raw
Materials branch.
c. Why is there fluctuation or dispersion in the oil? Due to the

fluctuation of the amount added to the mixture. We add to Oil the
smallest branch Quantity.
d. Why is there variation in the added amount of oil? Due to irregular

operation of the scale. The Balance branch is registered.
e. Why does the scale work irregularly? Why does it need

maintenance? In the Scale branch we place the Maintenance
branch.
Thus we continue expanding the Cause-Effect Diagram until it

contains all the possible causes of dispersion.
4. Finally we verify that all the factors that could cause dispersion have
been incorporated into the diagram. The Cause-Effect relationships
must be clearly established and in that case, the diagram is finished.
Let's look at an example from Kaoru Ishikawa's Quality Control Guide,
published by UNIPUB (No. York). This is a machine in which an
oscillating rotation defect occurs. The quality characteristic is the
oscillation of a shaft during rotation:

1 7
7
A Cause-Effect diagram is educational in itself; it helps people

understand in depth the process they work with, clearly visualizing the
relationships between the Effects and their Causes.
It also serves to guide discussions by clearly exposing the origins of a

quality problem. And it allows assignable causes to be found more
quickly when the process deviates from its usual operation.

1 7
7
Exercise: Cause - Effect Diagram

1 7
7
Example of: Cause - Effect Diagram

1 7
7
Healt Cheer Enthusias
h up m
o Relaxation Fighting spirit

m
• 7
O
Rest Nutrition Calories Patience
oc Devotion
oo
O'
z Fun Meal Careful
Calm
Time
Amount Concentration
Composure
Sleep
How well Trust

defeat in
a
Power tourname
Informatio Itinerar nt
n Motion y sports
Theor of the Exercis
Speed
y opponent Cooperation e
Planning _ .. , Quality
Analys quantity
, is 2 Teamwork
Rules Observation J
5P Advice
Common Judgment of the Functi —. Shape
sense situation on
Model '
Repetition
Match experience
Strategy Techni
que
APPLIED STATISTICS
Factors (Causes)
PRACTICE N° 6
Exercise: Cause - Effect Diagram
Delay in delivery of analysis results at the UMF #"XX".
a) Deficiency in equipment (cause) may be responsible for delayed

delivery of results.
b) The delay in analysis results (effect) may be due to the lack of an

adequate work plan.
c) Ignorance of how to perform some cultures (cause) may influence

the failure to deliver analysis results.
d) The delay in the delivery of analysis results (effect) is due to not

considering the period necessary for its completion (cause).
DISPERSION DIAGRAM
The investigation of a relationship between two variables begins with an

attempt to discover the approximate form of the relationship, for which
the observed data is represented in a coordinate system.
This graph is called a scatter diagram, which shows the location of the
values or points (Xi, y) of the two-dimensional variable (x, y), in a
rectangular coordinate system.
In the graph you can see whether or not there is a marked relationship
between the variables x and y, and you can see what shape it has: linear
or other.

17 7
9
Example 1: Below are the weights (x) and height (y) of 12 students.
Draw the scatterplot.
x 60 56 42 72 36 63 47 55 49 38 42 61
and 179 147 125 160 118 149 128 150 145 115
140 152
Solution: The corresponding scatter plot is shown in Figure 1
If we observe the position of the points in this scatter diagram, it is

evident that there is a linear dependence between the values of the
variables x and y, since as the weights increase, the height increases.

1 7
8
APPLIED STATISTICS
LINEAR CORRELATION
We have said that covariance measures the linear relationship between

x and y. To illustrate this, we draw the scatter diagrams in Fig. 1.
x x
b. Covariance close to zero
d. Covariance close to zero
Fig. 1 Different types of relationship between two variables and resulting covariance.
When x and y vary together linearly, as indicated in figures (a) and (c),
the covariance will be high. For example, in case (a) most of the Xi-Xeyi-
y points will be in the first quadrant and third quadrant (defined by x and
y), contributing positively to the sum. In case (c), most of the Xi-xeyi-y
points will be in the second and fourth quadrants, therefore they will be
negative, obtaining a high magnitude and negative sum in the
covariance formula.
On the contrary, when there is no relationship, case (b), or there is a

non-linear relationship, case (d), the variance will be small, as the points
are distributed across the four quadrants defined by x and y.
The drawback of covariance as a measure of linear association is its

dependence on the units of measurement of the variables.
Consequently, to construct a dimensionless measure, we will have to

divide the covariance by a term with the same dimensions. If it is divided
by the product of their standard deviations, the correlation coefficient
between two variables is defined.
Definition 1: The correlation coefficient (or Pearson linear correlation

18 7
1
index) between two variables x and y is defined by:

Cov (x, y)~
K
SxSy
where Sx and Sy are the standard deviations of x and y respectively.
It is shown that the correlation coefficient meets:
1. If x is multiplied by k i , and y by k2, the correlation coefficient does

not change.
2. If there is an exact linear relationship between both variables and all

points are on the line y = a + bx, the correlation coefficient is equal to i
(sib>0)or-1 (sib<0).
3. If there is no exact linear relationship: -1 < r < 1, which represents the

degree of association.
4. r = 0, when there is no linear correlation between x and y

Example 1: Two judges x and y of a contest separately assigned the
following scores to the 10 contest finalists.
X| 3 10 9 1246587
and | 10 8 10 4 1 2 5 6 7 9
1. Find the covariance given by the judges under the assumption that
they issued their judgment with absolute independence, thus
indicating the sign of dependence.
2. The correlation coefficient.
Solution 1. In this case the covariance is given by Z (x i -x)(y i -y) Z(xi-yi)

_
Cov (x, y)=------------n----------= —n— - xy

All the values we need are calculated in Table 1.
x i and i XiYi x i a 2
3 3 9 9 9 55
x
10 8 80 100 64 = 10 5.5.5
9 10 90 81 100
1
2
4
1
4
2
1
4
16
1
y-$8-55
4 2 8 16 4 Cov(x,y) = 372 - (5.5) (5.5)
6 5 30 36 25
5 6 30 25 36 = 37.2 - 30.25
8 7 56 64 49
7 9 63 49 81
= 6.95
55 55 372 385 385
This value indicates a direct positive dependence, since the
covariance also gives us the sign of the relationship.

1 7
8
APPLIED STATISTICS
3. From the table:

2 _ 2x i x 2 385 (5 5)2 ,ge
S x -~' X -10 99) = 8.25
s,e -M - y - 3 85 ■ (6.5y = 8.25
2
Therefore, the correlation coefficient is

Cov(x,y) _ 6.95 6.95 .
S x S y ^8.25 '.8.25 8.25 '
correlation with a high degree of association, since r varies from -1 to +1.
To see the degree of dependence we must consider r : r - (0.84)2 - 2 2
0.7056 (r is called the coefficient of determination) this indicates that

2
70.56% of the variation of y is explained by the variable x.

Misinterpretation of correlation coefficients
Sometimes the correlation coefficient is mistakenly interpreted as a

percentage. If r - 0.7, it will be incorrect to affirm that 70% of the
variations in y are explained by the variable x. For a correct
interpretation, must be considered. Thus for r - 0.7, it will be r - 0.49,
r2 2
only 49% of the variation of y is explained by the variable x.

CORRELATION COEFFICIENTS
SPEARMAN rank correlation coefficient
This correlation coefficient is used when one or both variables are

ordinal, that is, the elements of one or both series are order positions.
To define this correlation coefficient, the ranges of each variable are

considered; that is, the order numbers of the observations when they are
ordered in increasing or decreasing order. If there are two or more equal
values of the variable, the ranges of each of these are equal to the
average of the ranges that would correspond to them if they were
different consecutively.
The rank correlation coefficient "r" is calculated by the formula:
6 2d2
r s -1------=1----
n(n -1)
2
where d , - the difference of the ranges of x and y (d i - R x R y )
If one of the variables is not ordinal, a rank is assigned to the scores.

The interpretation of the Spearman rank correlation coefficient is the
same as that of the Pearson correlation coefficient.

18 7
3
APPLIED STATISTICS
Example 2: The attached Table shows the height of a father and son (in
inches). Find the Spearman correlation coefficient.
Solution: The heights of the parents and children, as well as the

corresponding ranks of the descending sorts, appear in the table .
x and d R and gave d2i

The ranges of values
father son equals are calculated like
65 68 9 6.5 2.5 6.25
for x:
63 66 11 10.5 0.5 0.25
67 68 7.5 6.5 1 1 5+6 11 —
64 65 10 12 -2 4
68 69 5.5 4 1.5 2.25 2 5255.5
62 66 12 10.5 1.5 2.25 7+8 = 15_ 7 5
71 68 2 6.5 -4.5 20.25 2 2
76 75 1 1 0 0
68 71 5.5 2 3.5 12.25 for and:
67 67 7.5 9 -1.5 2.25 5+6+7+8 26
69 68 4 6.5 -2.5 6.25
70 70 3 3 0 0 4 54565
57.00 10 + 11 _21 _e
-
2 2 - 10.5
Spearman's formula with n-12es
r=1
>62d 1-1
—1 GOES. •T -1 Q 1000 — Q o09
r -1
s
-T- - 19/1921N - 1 - 0.1980 - 0.802
2
n(n -1) 12(12 -1)
High degree of relationship.
Yule “Q” correlation coefficient.
It measures the association of two nominal qualitative variables with only

two exclusive categories: (sex: man, woman; studies: yes or no). When
the values of x and y are only two, the distribution of the number of
points can be arranged as the following table:
The correlation coefficient between this type of
By X 0 1 Total
\ variables x and y is given by.
0 to b a+b _ _ ad - bc
1 c d C+d u-
Total a+c b+d
ad + bc
The Q coefficient is not applicable when one of the boxes is zero.
Example 3: You want to study the relationship between finding a job and
knowledge of a certain language. For this purpose, a survey was carried
out, the distribution of the results is as follows:
Languag The value of the Yule
e Yea correlation coefficient
h No Total
Find a job, is: _ 11x10-7x4
Yeah 11 7 18
No 4 10 14 • 11x10 + 7x4
Total 15 17 32
= 82 = 0 594
-
138 -:994

18 7
4
PHI Correlation Coefficient It is a coefficient to measure the degree of
association between two ordinal or nominal qualitative variables, of only
two categories. The distribution of the number of points is arranged as in
the following table.
The PHIo correlation coefficient of points is

By 0 1 Total
XB given by
0 to b a+b ad-bc___________
__________________
1 c d c+d
Total a+c b+d n V(a+b)(c+d)(a+c)(b+d)
Example 4: see if there is a relationship between social activity and

classification as a leader .
Activate
Social
Qual. as a leader
high Low Total
Favorable 86 41 127
Unfavorable 62 71 133
Total 148 112 n = 260
Solution: The correlation coefficient © is :
= 86x71 -41 x62 = 6106-252 = 3564

Vl27x133x148x112 V279985216 16732.7588
= 0.213
Correlation and Causality
Correlation coefficients are only a measure of the covariation of the

variables.
As for the variation itself, it may be due to causes that affect each
variable in the same way or in opposite ways, or one of them is the
cause of the variation of the other, or it may happen that the causal
relationship is reciprocal; All this is unrelated to the verification of the
existence of the correlation and the value of the correlation coefficient.
It is easy for the learner to imagine pairs of variables that could give a
high value of a correlation coefficient and that is not really due to a close
relationship between them, but rather the common effect on them of a
third variable, and then this high value of the coefficient correlation
reflects only that common effect.
Consequently, the correlation coefficients must be handled very

carefully, since otherwise they may lead to erroneous conclusions.
SIMPLE LINEAR REGRESSION
The study of the association between two variables occurs in two

aspects. The first is known as correlation analysis, which is concerned
with determining "the degree of relationship" between the variables.

1 7
8
The second aspect is called regression analysis which tries to establish
the "form of the relationship between the variables"; That is, the
functional relationship between the variables is studied (y = f(x) or x =
f(y)), so that we can predict the value of one based on the other or
others. Conventionally, the variable or variables that are the basis of the
prediction are called the independent variable or variables and the
variable to be predicted is called the dependent variable.
The word regression was first used in this context by FRANCIS GALTON
(1822 - 1911) in his biological studies of heredity. They noted that the
average characteristics of the next generation of a particular group
tended to move in the direction of the average characteristics of the
population in general, rather than towards those of the previous
generation of that group. This trend was referred to as a regression
toward the population mean.
Then, regression has two meanings: one arises from the joint probability
distribution of two random variables; the second is empirical and arises
from the need to fit some function to a set of data.
The first meaning is outside the scope of this text. The second is more
practical. In it, the necessary elements are not available to determine the
regression curve; however, given a set of data, a functional form can be
assumed for the regression curve and an attempt may be made to fit it to
the data.
Thus, the regression of the variable y on x is given by;
y=f(x)
Definition 2: If the regression of y on x is a linear function, y = a + bx,
a, b constants
It is called simple linear regression of y on x. The value a is called the

ordinate at the origin, since it is the point where the straight line
intersects the y-axis. The slope of the line is measured by b, which gives
the change in y per unit change in the value of x.
The sign of b also indicates the type of relationship between x and y.

Estimation by the least squares method
A main task in regression analysis is to estimate the parameters a and b
of the line L: y = a + bx based on the sample data formed by the pairs (x
1 , y 1 ), (X 2 y 2 )... , (Xnyyn). The least squares method consists of
finding the line L, such that the sum of the squares of the differences of
the ordinates and. ya + bx., of the observed points (x¡, y) and the points
on the line L, (x ¡ , a + bx) be minimum (Fig. 1). That is, it is about finding
a and b in such a way that

1 7
8
n
SCD=2 [y i -(a + bx ¡ )J be minimum ¡ =1
Differentiating with respect to each of the parameters a and b, equating
to zero
is obtained:
eSD=2 2 Iy-(a+bx) (-1) = or
CSCD=2 2 Iy-(a+bx)1 (-Xi ) =

0
5b
Performing the respective operations in both equations, the system is

obtained:
2 y=b2x+ na
2 and xi = b 2x2+ aZx
Called normal equations for the least squares line
NOTE 5: If the data are tabulated, the expressions of the normal
equations will appear multiplied by their respective absolute frequencies.
Solving the system of normal equations for a and b we obtain

a = y-bx
_ 2 x i y i -nxy _ cov (x, y) /*
b
" 2X2-mx " v(x)
Then, the line L is written:
_ Cov (x, y) . y=y + v(x) (xx)
Call least squares line from y to x.

The least squares line L' of x in y is obtained by exchanging the roles of
x and y in the least squares method:
L':x = x + b'(y - y), where b'= Cov (x,y)/v(y)
or also:
YsYb(X- x)
If the linear fit is correct, we must have 1/b'= b, that is, bb'= 1
Note that the coefficient of determination is written:
2 _ Cov (x,y) Cov (x,y) _ Cov (x,y) Cov (x, y) _ r " S x S y S x S y "
V(x) V(y) ■ bb
Then the correlation coefficient is r - bb'. If b > 0, it is the positive root;
but if b < 0, r is the negative square root. Then, the sign of r indicates the
direction of the relationship between x and y. Therefore, if r 2 approaches
1 the linear fit is valid.
PARETO CHART
Concept
The Pareto Diagram is a graph where various classifications of data are
organized in descending order, from left to right by means of simple bars
1 7
8
after having gathered the data to qualify the causes. So that an order of
priorities can be assigned.
What is it?
The name Pareto was given by Dr. Joseph Juran in honor of the Italian
economist Vilfredo Pareto (1848-1923) who conducted a study on the
distribution of wealth, in which he discovered that the minority of the
population owned the majority of wealth and the majority of the
population owned the least of the wealth. With this he established the so-
called "Pareto's Law" according to which economic inequality is
inevitable in any society.
Dr. Juran applied this concept to quality, obtaining what is known today
as the 80/20 rule.
According to this concept, if you have a problem with many causes, we
can say that 20% of the causes solve 80% of the problems! problem and
80% of the causes only solve 20% of the problem.
Therefore, Pareto Analysis is a technique that separates the "vital few"
from the "trivial many." A Pareto chart is used to graphically separate the
significant aspects of a problem from the trivial ones so that a team
knows where to direct its efforts to improve. Reducing the most
significant problems (the longest bars on a Pareto Chart) will do more for
an overall improvement than reducing the smaller ones. Often one
aspect will have 80% of the problems.
In the rest of the cases, between 2 and 3 aspects will be responsible for
80% of the problems.
In relation to the management styles of Problem Solving and Decision
Making (Conservative, Firefighter, Opportunist and Integrator) 1[1], we
see how the use of this tool can be an excellent alternative for a
Firefighter style manager, who constantly When solving problems he
only "puts out fires" that is, he puts all his effort into the "trivial many."
When it's used?
• When identifying a product or service for analysis to improve quality.
• When there is a need to draw attention to problems or causes in a
systematic way.
• By identifying opportunities to improve.
• By analyzing the different groupings of data (e.g. by product by
market segment, geographic area, etc.).
• By looking for root causes of problems and prioritizing solutions.
• When evaluating the results of the changes made to a process

(before and after).
• When the data can be classified into categories.
• When the rank of each category is important.

Pareto is a widely used data analysis tool and is therefore useful in

1 7
8
determining root cause during a problem-solving effort. This allows you
to see what the biggest problems are, allowing groups to establish
priorities. In typical cases, the few (steps, services, items, problems,
causes) are responsible for the
most of the negative impact on quality. If we focus our attention on these
vital few, we can make the most potential profit.
of our efforts to improve quality.
A team can use the Pareto Chart for several purposes during a project to
achieve improvements:
• To analyze the causes
• To study the results
• To plan continuous improvement.
•Pareto Charts are especially valuable as “before and after” photos to
demonstrate what progress has been made. As such, the Pareto Chart
is a simple but powerful tool.
How is it used?
1. Select logical categories for the time period topic).
2. Gather data. The use of a Check List can be very helpful in this step.
3. Sort the data from the largest category to the smallest.
4. Total the data for all categories.
5. Calculate the percentage of the total that each category represents.
6. Plot the horizontal (x) and vertical (and primary and secondary)
axes
7. Plot the left vertical axis scale for frequency (from 0 to total), as
calculated above.
8. From left to right plot the bars for each category in descending order.
If there is an “other” category, it should be placed at the end,
regardless of its value. That is, it should not be taken into account
when ordering the frequency of the categories from highest to lowest.
9. Plot the right vertical axis scale for the cumulative percentage,
starting at 0 and going up to 100%
10. Plot the line graph for the cumulative percentage, starting at the top
of the first (highest) category bar.
11. Give the graph a title, add the dates when the data was collected,
and cite the source of the data.
12. Analyze the graph to determine the "vital few"
Tips for construction / interpretation
As we have seen, a Pareto Chart is a bar chart that lists categories in
descending order from left to right, which can be used by a team to
analyze causes, study results and plan continuous improvement.

1 7
8
Among the difficulties that may arise when trying to interpret the Pareto
Diagram is that sometimes the data does not indicate a clear distinction
between the categories. This can be seen on the graph when all the bars
are more or less the same height.
Another difficulty is that more than half of the categories are needed to
add more than 60% of the quality effect, so a good analysis and
interpretation depends largely on a good prior analysis of the causes and
subsequent data collection. .
In either case, it seems that the Pareto principle does not apply.
Because it has been proven valid in literally thousands of situations, it is
very unlikely that an exception will have been found. It is much more
likely that an appropriate breakdown of the categories was simply not
selected. You should try to stratify the data in a different way and repeat
the Pareto Analysis.
This leads us to the conclusion that to carry out a Problem
Solving/Decision Making (RP/TD) process it is necessary to handle each
of the basic quality tools, both from a theoretical point of view and from
its application. .
The interpretation of a Pareto Chart can be defined by completing the
following example sentences:
"There are (number of categories) contributors related to (effect). But
these (number of vital few) correspond to the (number) % of the total
(effect). We must pursue these vital few (number) categories, as they
represent the greatest potential gain for our efforts.”
Relationship with other tools
A Pareto Diagram generally relates to:
• Cause and Effect Diagram (Ishikawa)

• Review Check List
• Data Collection Check List
• Action Planning Matrix
Application example
A refrigerator manufacturer wants to analyze what are the most frequent

defects that appear in the units when they leave the production line. For
this, he began by classifying all possible defects into their various types:
Type of deffect Problem Detail

Engine does not stop Does not stop the engine when it reaches
temperature
does not cool The motor starts but the ice cream parlor does
Def. not cool
Broken or deformed weather stripping that does
Paint Def. Paint defects on external surfaces
Stripes Stripes on external surfaces
It does not work When plugged in the engine does not start
Door does not close The door does not close properly
Drawers Def. Interior drawers with cracks
Engine does not start Engine does not start after stop cycle
Bad leveling The refrigerator rocks and cannot be leveled

1 7
9
Door Def. Refrigerator door does not close tightly
Others Other defects not included in the above
Subsequently, an inspector checks each refrigerator as it leaves

production, recording its defects according to these types. After
inspecting 88 refrigerators, a table like this was obtained:
Type of deffect Problem Detail Freq.

Def. Broken or deformed weather stripping that does 9
not fit
Paint Def. Paint defects on external surfaces 5
Drawers Def. Interior drawers with cracks 1
Bad Leveling The refrigerator rocks and cannot be leveled 1
Engine does not Engine does not start after stop cycle 1
start
Engine does not Does not stop the engine when it reaches 36
stop not cool
Does temperature
The motor starts but the refrigerator does not cool 27
It does not work When plugged in the engine does not start 2
Others Other defects not included in the above 0
Door Def. Refrigerator door does not close tightly 0
Door does not close The door does not close properly 2
Stripes Stripes on external surfaces 4
Total: 88

1 7
9
The last column shows the number of refrigerators that had each type of
defect, that is, the frequency with which each defect occurs. Instead of
the numerical frequency we can use the percentage frequency, that is,
the percentage of refrigerators in each type of defect:
Type of deffect Problem Detail Freq. Freq. %

Broken or deformed weather stripping that does
Def. 9 10.2
not fit
Paint Def. Paint defects on external surfaces 5 5.7
Drawers Def. Interior drawers with cracks 1 1.1
Bad Leveling The refrigerator rocks and cannot be leveled 1 1.1
Engine does not Engine does not start after stop cycle 1 1.1
start
Engine does not Does not stop the engine when it reaches 36 40.9
stop
Does not cool temperature
The motor starts but the refrigerator does not cool 27 30.7
It does not work When plugged in the engine does not start 2 2.3
Others Other defects not included in the above 0 0.0
Door Def. Refrigerator door does not close tightly 0 0.0
Door does not The door does not close properly 2 2.3
close
Stripes Stripes on external surfaces 4 4.5
Total 88 100
We can now represent the data in a histogram like the following:
Frequency of defects found
But what are the defects that appear most frequently? To make it more
obvious, before graphing we can sort the table data in decreasing order
of frequency:

1 7
9
Type of
Problem Detail Freq. Freq. %
deffect
Engine does not Does not stop the engine when it reaches 36 40.9
stop temperature
The motor starts but the refrigerator does not
Does not cool 27 30.7
Def. cool
Broken or deformed weather stripping that does 9 10.2
Paint Def. Paint defects on external surfaces 5 5.7
Stripes Stripes on external surfaces 4 4.5
It does not work When plugged in the engine does not start 2 2.3
Door does not close The door does not close properly 2 2.
Drawers Def. Interior drawers with cracks 1 1.1
Bad Leveling The refrigerator rocks and cannot be leveled 1 1.1
Engine does not Engine does not start after stop cycle 1 1.1
start
Door Def. Refrigerator door does not close tightly 0 0.0
Others Other defects not included in the above. 0 0.0
Total: 88 100
We see that the "other" category should always go last, regardless of its
value. In this way, if it had had a higher value, it should still have been
placed in the last row.
Now it is evident which are the most frequent types of defects. We can
see that the first 3 types of defects occur in approximately 82% of
refrigerators.
By the Pareto Principle, we conclude that: Most of the defects found in

the batch belong to only 3 types of defects, so that if the causes that
cause them are eliminated, most of the defects would disappear.

1 7
9
PRACTICE N° 7
1. Pareto chart
2. Cause - Effect Diagram
3. Flowchart
4. Control Diagram
Exercise: Pareto Diagram
In the Systems Department there are work stoppages due to failures of
certain machines. It is decided to analyze this problem to make decisions
aimed at solving it.
The area staff lists the main causes that may be influencing the problem.
• Electrical power interruption.
• Incorrect handling by the operator.
• Inadequate program
• Lack of maintenance
• Viruses in the system
• Others
To confirm such observations, it was decided to collect data in any of the
two shifts (morning-afternoon) randomly for 8 hours, on each of the four
days determined as the collection period.
Organized:
Work stoppage
Work Stoppage Time (in Accumulated

Causes Relative % of downtime
minutes) (f) Relative %
Lack of maintenance 202

Inadequate program 114
Electrical power 92
interruption
Incorrect management of 45
learning
Viruses in the system 19
Others 16
Total 488
Getting relative percentage

% frequency of the cause X 100
total frequency
Construction of the diagram and determination of “vital few”
100 %
80%
60%
40%
20%
CONTROL DIAGRAM
Concept:
The process of determining what is being carried out, in order to
establish the necessary corrective measures and thus avoid deviations
in the execution of the plans.
Since control implies the existence of goals and plans, no manager can
19
4
control without them. He cannot measure whether his subordinates are
operating in the desired manner unless he has a plan, whether short,
medium or long term. Generally, the clearer, more complete, and
coordinated the plans are and the longer the period they cover, the more
complete the control can be.
A manager can study past plans to see where and how they went wrong,
to discover what happened and why, and take steps to prevent errors
from happening again. However, the best control prevents deviations
from happening, anticipating them.
Control elements.
1. Relationship with what was planned. Control always exists to verify
the achievement of the objectives established in planning.
2. Measurement. To control, it is essential to measure and quantify
the results.
3. Detect deviations. One of the functions inherent to control is to
discover the differences that arise between execution and planning.
4. Establish corrective measures. The object of control is to foresee
and correct errors.
2. Control sheet
The Control Sheet or data collection sheet, also called Record, serves
to gather and classify information according to certain categories, by
noting and recording its frequencies in the form of data. Once the
phenomenon that needs to be studied has been established and the
categories that characterize them have been identified, these are
recorded on a sheet, indicating the frequency of observation.
The essential thing about data is that the purpose is clear and that the
data reflects the truth. These collection sheets have many functions,
but the main one is to make data collection easy and done in a way
that it can be easily used and analyzed automatically.
In general, data collection sheets have the following functions:
Distribution of variations of variables of the articles produced (weight,
volume, length, size, class, quality, etc...)
• Defective item classification
• Locating defects in parts.
• Of causes of defects.
• Verification check or maintenance tasks.
Once the reasons for collecting data have been established, it is

important to analyze the following questions:
• The information is qualitative or quantitative
• How the data will be collected and in what type of document it
will be done.
• How the information collected is used
• How to analyze
• Who will be in charge of data collection
• How often will it be analyzed?
• Where is it going to take place?
1 7
9
This is a manual tool, in which data is classified through marks on the
reading made instead of writing them. For these purposes, some printed
formats are used. The most important objectives of the control sheet are:
• Investigate distribution processes
• Defective items
• Defect localization
• Causes of effects
A sequence of useful steps to apply this sheet in a Workshop is as
follows:
1. Identify the tracking element
2. Define the scope of the data to be collected
3. Set the periodicity of the data to be collected
4.Design the format of the data collection sheet, according to the
amount of information to be collected, leaving a space to total the
data, which allows knowing the start and end dates, the probable
interruptions, the person who collects the information. , sources,
etc.
Scatter plots
They are used to study the variation of a process and determine what
causes this variation.
A control chart is a linear graph in which an upper limit (upper control
limit) and a lower limit (lower control limit) have been statistically
determined on either side of the mean or center line. The center line
reflects the product of the process. Control limits provide statistical
signals for management to act, indicating the separation between
common variation and special variation.
These charts are very useful for studying product properties, process
variables, costs, errors and other administrative data.

1 7
9
A control chart shows
1. If a process is under control, it is not

2. Indicates results that require explanation
3. It defines the capacity limits of the system, which after comparison
with the specification limits can determine the next steps in an
improvement process.
This can be a broken line or a circle. The broken line is often used to
indicate dynamic changes. The broken line is the control chart that
provides information on the status of a process and indicates whether
the process is established or not. Example of a control chart, where the
proposed measures versus time.
It clarifies how the measurements are related to the upper and lower
control limits of the process, the points outside the control limits show
that the control is out of control.
All quality controls require a certain sense of judgment and own actions
based on information collected in the workplace. Quality cannot be
achieved solely through calculations developed on the desk, but through
activities carried out in the plant and based, of course, on desk
calculations.
Quality control or quality assurance began with the idea of emphasizing
inspection.
Need for full participation
To apply quality assurance from the beginning in the development stage
of a new product, it will be necessary for all divisions of the company and
all its employees to participate in quality control.
When quality control only emphasizes inspection, only one division is
involved, either the inspection division or the quality control division, and
this is limited to checking at the exit door to prevent defective products
from leaving.
However, the quality control program emphasizes the manufacturing
process; participation extends to assembly lines, subcontractors, and
purchasing, product engineering, and marketing divisions.
In a more advanced application of quality control, which is the third
phase, all of the above is considered insufficient. Participation already
has to be on a company-wide scale.
This means that those involved in planning, design and research of new
products, as well as those in the manufacturing division and in the
accounting, personnel and labor relations divisions, have to participate
without exception.
Quality assurance has to reach this third phase of development, which is
the application of quality assurance from the early stages of a product's
development. At the same time, quality control has embraced the
concept of total participation by all divisions and their employees. The
convergence of these two trends has given rise to enterprise-wide quality
control, the most important feature of Japanese Quality Control today.
UCTION MECHANICS 142/147
1
9
In the manufacture of high-quality products with full quality guarantee,
we must not forget the role of workers. The workers are the ones who
produce, and if they and their supervisors do not do it well, Quality
Control will not be able to progress.
The satisfaction of a job well done with quality. This includes the
following:
The joy of completing a project or achieving a goal
The joy of climbing a mountain simply because it is there.
It is suggested that specialized manufacturers be established in their
own fields, at least in each province. Otherwise we will not be able to
improve quality or increase productivity.
Exercise: Control Diagram
Example:
In the process of washing hospital clothing (sheets) it was seen the
convenience of controlling the acidity of the disinfectant (PH), to avoid
deterioration of the garment without affecting the level of asepsis in the
clothing.
For this purpose, tests were done every 20 minutes and the results were
recorded in groups of size 6. The measurements are presented in the
following table.
MEASUREM MEASUREMENT GROUPS (20')
DAY ENT “K” x R
NUMBER to b c d and F 1 .- Calculate averages for
2/8 1 4.22 4.20 4.20 4.08 4.15 4.15 4.166 0.14 each of the k subgroups
_ X 1 + X 2 + ... xk
2/8 2 4.25 4.22 4.10 4.15 4.10 4.15 4.161 0.15 neither
2/8 3 4.20 4.22 4.01 4.17 4.18 4.30 4.180 0.29

2 .- Calculate the rank of each
2/8 4 4.20 4.25 4.18 4.05 4.17 4.10 4.158 0.20 subgroup (k)
3/8 5 4.15 4.20 4.03 4.15 4.31 4.25 4.181 0.28
3 .-
3/8 6 4.18 4.17 4.17 4.30 4.15 4.10 4.178 0.20 LR i R1 + R2 + R3 + ... Rk
3/8 7 4.22 4.24 4.23 4.07 4.19 4.25 4.196 0.18 R
k k
3/8 8 4.00 4.15 4.18 4.10 4.30 4.20 4.155 0.30
4 . Range graphs D3 and D4
4/8 9 4.10 4.15 4.30 4.36 4.15 4.10 4.193 0.26 are constant
4/8 10 4.20 4.25 4.20 4.30 4.15 4.10 4.200 0.20 LICR = D3 R
4/8 11 4.35 4.20 4.25 4.12 4.30 4.10 4.220 0.25 LICR =
LSCR = D4 R
45.988 245
_ X1+X2+ ... Xk
X=
------------k---------
x=
LICX = x-A2 R LICx =

LICX = LSCX = LSCx = LSCx =
0.
0.—-------1------l
2
Subgroups
pH
5.
INTERPRETATION: The
4.10-.
process shows
1
1 4 16 18 1 ।
PARTIAL
0.-
2 1
EVALUATION II
0. TO
LSCR =
1.
pH
0.222 4. ---------------------------------------------------------------------------------------------------------------- 5. Average graphs (x)
4.2EIT
—■
HER
The range measures the amplitude of the sample values and is
calculated by VF as the difference between the highest and lowest
value.
2. The variance measures the distance between the VF values of the

19
8
series and the mean.
Introduction............................................................................................2
2. Division of statistics........................................................................8
2.1 Descriptive statistics....................................................................8
85+90+ 93+82+95 = 415 = 89 points....................................................9
Issues:...................................................................................................11
Solution:...............................................................................................15
Solution:...............................................................................................16
1. Data collection..............................................................................18
Example:..............................................................................................22
Solution:...............................................................................................23
PRACTICE N° 2..................................................................................25
1.1 Introduction...............................................................................32
Concept:...............................................................................................33

1 7
9
consciousness)......................................................................................40
ll.Quality...............................................................................................40
QA........................................................................................................43
QA........................................................................................................44
1. "Control is not classification or selection"....................................45
allows analysis."...................................................................................45
QA........................................................................................................47
Technical rules.....................................................................................53
ISO 9000 STANDARDS.....................................................................54
ISO 9000 objectives.............................................................................55
20
0
ISO 9000 standards..............................................................................55
STANDARD IS014000.......................................................................55
IS014000 Standards.............................................................................56
Solution:...............................................................................................57
Ni=n+n2+na + ... + n i.............................................................................58
Interpretation:.......................................................................................60
Solution................................................................................................66
Lower limit Upper limit.....................................................................70
Observations:........................................................................................76
Y1-y'i=<y'i-1 , y'i].................................................................77
and. — 93.5494.5 — 94.0.......................................................................79
Solution to:...........................................................................................83
+68......................................................................................................................97
PRACTICE N°3.................................................................................103
AVERAGE.........................................................................................106
Observation........................................................................................106
> (X ¡ -X) = (X I -x) + ... + (X n -X) = (X I + ... +X
n )-n x = n x -n x = 0............................................................106
2x - X|20..............................................................................................107
Example..............................................................................................107
Solution..............................................................................................107
ge1/gn240_24............................................................................................107
> 2(xx)2.........................................................................................................114
Observation........................................................................................114
The mean square................................................................................121
THE MEDIAN...................................................................................122
2 7
0
Observation........................................................................................122
Observation........................................................................................122
FASHION..........................................................................................124
Observation........................................................................................125
EXAMPLE.........................................................................................126
PARTIAL EVALUATION I..............................................................127
RANGE, VARIANCE, STANDARD DEVIATION.........................144
= (0,010)= 0,0320.......................................................................145
Breakdown.........................................................................................146
Where:................................................................................................146
Example..............................................................................................146
X=1 2x..............................................................................................................147
a = 12(x-7)2..........................................................................................147
°\4..................................................................................................................147
°s\4............................................................................................................147
PRACTICE N° 4................................................................................148
from $10, from..............................................................................................148
C ¡ xn =ni(ocxh= h) C ¡ Ci................................................149
Solution:.............................................................................................155
Line graphs.........................................................................................158
PRACTICE N° 5..................................................................................159
1. Concept:......................................................................................165
2. Application:................................................................................165
5. Typing.........................................................................................167
_x-u........................................................................................................167
-..............................................................................................................167
— —- z.................................................................................................167
Z
= npq is N(0,1)...........................................................................................168
............................................................................................................168
Example:............................................................................................171
20
2
Concept:.............................................................................................173
DISPERSION DIAGRAM.................................................................180
> 6 2 d............................................................................................................185
-
138 -:994........................................................................................185
V(a+b)(c+d)(a+c)(b+d).......................................................................186
CSCD=2 2 Iy-(a+bx)1 (-Xi ) = 0........................................................188
5b...........................................................................................................188
2 and xi = b 2x2+ aZx.............................................................188
YsYb(X- x)................................................................................................188
Concept..............................................................................................188
What is it?..........................................................................................189
When it's used?...................................................................................189
How is it used?...................................................................................190
PRACTICE N° 7................................................................................195
Organized:..........................................................................................195
Work stoppage...................................................................................195
Concept:.............................................................................................195
2. Control sheet..................................................................................196
Scatter plots........................................................................................197
A control chart shows.........................................................................198

2 7
0
First Step: Defining the Problem........................................................205
Activities:...........................................................................................207
(Conclusions).....................................................................................208
Activities............................................................................................208
3.
THE QUALITY ROUTE
The Quality Path is a standard problem-solving procedure. This is a kind

of count or representation of the activities related to the Quality Control
Cycle: Plan, Do, Check, Act (PHVA). It consists of the following seven
steps:
a) Definition of the problem.

b) Recognition of the Characteristics of the Problem (Observation).
c) Search for the main causes (Analysis).
d) Actions to eliminate the causes (Action).
e) Confirmation of the effectiveness of the action (Verification).
f) Permanent elimination of causes (Standardization).
g) Review of activities and planning of future work (Conclusions).
The first three steps correspond to the Plan action, the fourth step to the
Do action, the fifth step to the Verify action and the sixth step to the Act
action, of the Quality Control Circle. With step seven this Control Circle
begins again.
First Step: Defining the Problem.
This is the first step of the standardized problem-solving procedure or

Quality Route. We must begin by defining what we mean by a problem.
For our purposes, the problem is defined as the undesired result of a job,
the deviation from a standard or a functioning norm, or the deviation from
what should be.
In this sense: Customer non-satisfaction, results that do not agree with

the objectives or goals or anything that deviates from the policies,
represent problems for an organization.
Analyzing this definition we can see that to define a problem it is first
necessary to know what is desirable, what the Client wants, in two words
the "should be".

20
4
This leads us to recognize the importance of the objectives and the
commitment of managers or Senior Management to define the
objectives.
It will allow themselves, the Middle Management, the Supervisors and
the Operatives to know the direction of the company and in this way
define its problems.
The activities that must be carried out in this first step are:
a) Knowledge of the guidelines, objectives and goals of the organization

or area of study.
b) Identification of priority problems, comparing the results obtained with

what was planned. To do this, histograms, control charts or various
graphs can be used, as well as the Pareto Diagram.
It is advisable to use a Problem Selection Matrix, a technique that is

currently widely used to assess and prioritize problems based on factors
such as: importance, frequency, cost, accessibility, among others.
c) Selection of a problem among all the many problems that have been
identified. The choice of this problem must be based on its importance
(it must be much more important than any other) and the objective of
improvement: quality, availability, safety, work environment, service,
etc.
d) Definition of those responsible for solving the problem. It can be one

person, a team of people such as an Improvement Team or a Quality
Circle.
e) Preparation of a budget for improvement and a schedule of activities

(Gant Chart) that allows us to plan what we want or need based on
time.

(Observation) Activities
a) Analysis and understanding of the problem. The time, place and

context where the problem occurs as well as the many points of view
must be investigated to discover the variation in the result. At this
point, the determination of certain measurement indicators of the
problem is required in order to have a more objective explanation or
evidence.
b) Setting a quantitative goal of what is desired, based on the previous

point
The tools typically used in this step are the Pareto Chart and the
Control Chart.
Third Step: Search for the Main Causes (Analysis) Activities
a) Thorough analysis of all the possible causes that may cause the
problem, with the participation of all the people involved in the
problem. That is to say, what is called the hypotheses of causes are
proposed.
2 7
0
To do this, a cause-effect diagram must be made, using the
information obtained from observation. From this Diagram determine
the causes that seem to have a high priority of being the main ones.
b) Test the most probable causes (hypotheses of causes), in order to

verify and conclude with the determination of the causes that really
have an impact on the problem.
This sometimes requires new information or new experiments. The tool

used to verify the causes is basically the Data Collection Sheet. It is also
recommended to apply surveys or another tool within a carefully
designed plan.
Fourth Step: Actions to eliminate the causes (Action) Activities
a) Approach to alternative solutions to eliminate the causes of the
problem. It is necessary to distinguish here the solutions that only
constitute immediate remedies from those that actually eliminate the
causal factors. The advantages and disadvantages of each designed
alternative must be examined, selecting the one that is most
convenient.
b) Design measures for side effects, if necessary. In addition to the tools
exposed in this step, the Diagram called "HOW" is usually used to
formulate the solution alternatives, then in the Gant Chart to program
the implementation.
(Verification). Activities
a) Comparison of the results obtained with the implemented solution with
those obtained previously, using histograms, linear graphs, control
graphs or any other graph that is useful for this purpose.
b) Measurement of the effect in monetary terms and compare with the
desired objective.
This is a typical phase of monitoring the implemented improvements.
(Standardization)
Activities:
a) Formalization of the new standards that reflect the improvement in
manuals of: operation, procedures, specifications of new control
limits, etc.
b) Communication of the new standards to all those involved.
c) Training and training of personnel.
d) Design of a monitoring system to verify the application of the new
standards.
(Conclusions).
Activities
a) Review of everything carried out, benefits obtained, experiments
carried out, difficulties encountered, degree of participation of the
people involved, costs incurred, tools used, etc.
b) Preparation of a list of unresolved problems, including new problems
that have arisen.

20
6
Definition of the new problem to be solved, and continuing indefinitely
with the quality improvement process.

2 7
0
BIBLIOGRAPHY
• INTRODUCTION TO PROBABILITY THEORY

HAROLD J, LARSON
• THE TEACHING OF STATISTICS
ETAYO MIQUEO, JOSÉ

• ALPHA 10
NORMA EDITORIAL
• WHAT IS TOTAL QUALITY CONTROL?
KAORU ISHIKAWA
• TOTAL QUALITY
EDITORIAL NORMA
• TOTAL QUALITY AND LOGISTICS
PAX EDITORIAL
• INTRODUCTION TO QUALITY MANAGEMENT
SOLER GARCÍA, DAVID
CAMPINS MASRIERA,
• WWW.MONOGRAFIAS.COM/TRABAJOS15/
JUAN
TOTAL QUALITY/TOTAL-QUALITY.SHTML
• WWW.AITECO.COM/CALIDAD.HTM
INTERNET
• WWW.IMPROVEN-CONSULTAS.COM
INTERNET
INTERNET
2
0
SENATI INTELLECTUAL PROPERTY PROHIBITED
ITS REPRODUCTION AND SALE WITHOUT AUTHORIZATION
CORRESPONDENT
MATERIAL CODE 0751 EDITION

APRIL 2006
2. Variance: remember that the mean of this sample is 1.253. Then,
we apply the formula:

Applied Statistics

Uploaded by

Copyright:

Available Formats

Applied Statistics

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applied Statistics

Uploaded by

Copyright:

Available Formats

National Industrial Work Training Service

Operational Level Technician

The evolution and development of "Statistics" in today's world is such that it

1. Definition: Statistics is a science that provides a set of methods used to

The field of Statistics is generally divided into two large areas.

2.1 Descriptive statistics

PRODUCTION MECHANICS 10/14

describing it, is what constitutes Descriptive Statistics.

Average or arithmetic mean, which is the sum of the observed values

85+90+ 93+82+95 = 415 = 89 points

The second aspect of the definition of Statistics is: predicting

PRODUCTION MECHANICS 10/14

Now this prediction is made with a certain degree of confidence; This

Since it is impossible to test the vaccine on all people, let us consider

The problem can also be stated as follows: The proportion of non-

PRODUCTION MECHANICS 3/14

c. The difference between the two averages is 20 hours.

f. You decide to buy brand A bulbs instead of brand B bulbs.

a. The average duration of the four Aes brand bulbs

1100 + 900 + 980 + 1020 _ 4000 _ .000 is

1050 + 960 + 1065 + 845 + 980 _ 4900 _ ho

PRODUCTION MECHANICS 10/14

a. Juana's average grade is 16 and Juan's average grade is 14.

b. Juana is a better student than Juan.

d. The difference between the two averages is 2 points.

4. In what context is the word "statistics" generally used on radio than on

5. Classify each of the following statements as either inferences or descriptive

c. The company "RM" predicted who would be the winner in a presidential

The concept of "population", statistically speaking, is a broader concept than

PRODUCTION MECHANICS 10/14

Population from a statistical point of view is defined as follows.

Population: It is the collection of all individuals, objects u

The terms population and universe are often used interchangeably.

a. The ages of students in Peru.

In each of the previous examples, there is at least one common characteristic

When referring to the nature of the problem, we also choose to distinguish

(b) . Any population can be real or virtual.

PRODUCTION MECHANICS 10/14

Notation: We will denote the size of the finite population by "N".

Parameter: It is a summary measure that describes a characteristic of the

- The population mean = u (read mn)

Obviously it is impossible to calculate the true value of any parameter of an

Of course, the number of observations in a sample is less than the number of

For example, if you wanted to estimate the average annual expenditure of

PRODUCTION MECHANICS 10/14

number of students, then determine the annual expenditure corresponding to

Based on the sample average (sample average) obtained in this way, an

The population proportion, which is an unknown parameter, is the proportion of

For example, Cl (Intelligence quotient) average of a sample of first-grade

PRODUCTION MECHANICS 10/14

Example: A preliminary vote is taken to determine the preferences of voters in

a. What constitutes the sample?

a. The sample consists of the responses of 1,500 registered voters.

b. The population is made up of the responses of all registered voters.

d. The population parameter will be the proportion of all registered voters

Review exercises related to basic descriptive statistics

1. During a certain week, a restaurant served 2,300 customers. Explain the

a. What is the proportion of voters in the sample who belong to party A?

PRODUCTION MECHANICS 10/14

Name Civil status ________