Fmea, RCM

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 10

FMEA/FMECA

Failure Modes and Effects (Criticality) Analysis


Glossary
CA = Criticality Analysis
CAR = Corrective Action Request
CC = Contributing Cause
DC = Direct Cause
FBD = Functional Block Diagram
FMEA = Failure Modes and Effects Analysis
FMECA = Failure Modes and Effects Criticality Analysis
FR = Failure Rate
FS = Failure Scenario
FTA = Fault Tree Analysis
MTBF = mean time between failures
NTF = No Fault Found
OSD = Occurrence, Severity, Detectability
PCA = Preventive Corrective Action
PFM = Potential Failure Mode
PIPs = Performance Improvement Projects
QLA = Qualitative Assessment
QTA = Quantitative Assessment
RBA = Risk-Based Approach
RC = Root Cause
RCA = Root Cause Analysis
RPN = Risk Priority Number
RRA = Revised Risk Analysis
SCA = Specific Corrective Actions
SWIFT = Structured WHAT IF Technique
SYCA = Systematic Corrective Actions
RBD = Reliability Block Diagram

FMEA is a methodology that helps to identify potential failure modes and


their effects(consequences) and implement a strategy to prevent or lower the
1

failures and also the risks. It is used for a produc life-cycle and its based on
experience from similar products/processes or based on failure logic.
The FMEA can be done without a CA, but a CA requires the FMEA has
already identified system level critical failures. Only after this, the result is
FMECA.
EFFECTS analysis refers to analysing the consequences of failure based on
a system levels structure.
RISK REDUCTION main goal is to reduce the severity of failure effects or
to lower the probability of failure, or both. The probability should get to
impossible to occur, fact that can happen by eliminating the root cause(s). It is
ideally to early identify critical failure modes so they can be eliminated or at least
minimized.
FMECA is structured in 2 sub-analyses:
1. FAILURE MODES (FM)

EFFECTS ANALYSIS (EA)

2. CRITICALITY ANALYSIS (CA)


FAULT TREE ANALYSIS (FTA) is a top-down deductive failure analysis
which may handle multiple failures, using a backward logic. It is used BOOLEAN
LOGIC.

The GATE symbols describe the relationship between input and output events.

BENEFITS
A well implemented FMEA should provide:

High probability of successful operation and safety


Knowledge of potential failure modes and their impact on system operation
A list of failure modes ranked, according to the severity and probability of
occurrence.
Early planning of tests.

The failure MODE is a specific manner by which a failure occurs.


The MECHANISM (PROCESS) is a secquence of causes that leads to a failure
mode, over a period of time. (Ex: fatigue,corrosion)

DORMANCY/LATENCY PERIOD
Is the average time that a failure mode may be undetected:

Seconds auto detected by maintenance computer


8 hours detected by turn-around inspection
2 months detected by scheduled maintenance
2 years detected by overhaul task (revizie generala)

RISK PRIORITY NUMBER (RPN) = SEVERITY(S) x OCCURRENCE(O) x


DETECTION(D)
RPN is a product of the numerical Severity, Occurrence and Detection
ratings.
Risk is a measurement of consequence of a failure mode related to its probability
of occurrence. High risk failure modes should receive risk reduction
considerations.

FAILURE CAUSES
FAILURE CAUSES are defects in design, process and quality which are the
cause of a failure or which initiate a process which leads to a failure.
FAILURE MECHANISM : corrosion, welding, fatigue, dust accumulation
ROOT CAUSE: human error design failure, operational errors,
management failure,
specification failure.
FAILURE SCENARIO : secquence, combination of
events/failures/conditions leading to an
end failure state.
- A failure scenario is for a system the same
as the failure mechanism is for a component.
- Scenario/Mechanism = complete
description of the failure occurrence.
Mechanical failure:

Excessive DEFLECTION (deviatie, abatere)


3

BUCKLING (deformare)
DUCTILE FRACTURE (fisura)
BRITTLE FRACTURE (casare)
IMPACT
CREEP (alungire, dilatare)
RELAXATION (slabire)
THERMAL SHOCK
CORROSION
WEAR (frecare)
FATIGUE

ROOT CAUSE
ROOT CAUSE is the initial/initiating cause that leads to a failure mode. It
describes the depth in the causal chain.
Root = deep, basic, fundamental, underlying, initial

ROOT CAUSE ANALYSIS (RCA)


RCA is a method of problem solving used to identify the root cause of a
failure. A factor is considered a ROOT CAUSE if its removal prevents the final
potential failure.
The main goal is to :
Prevent recurrence
Be effective
Have a secquence of events for understanding the relationship
betweet factors.
--General process for performing and documenting an RCA-based Corrective
Action-RCA forms the most critical part of successful corrective action. So the steps we
have to proceed are the following:
1. Define the problem, describe the event including qualitative and
quantitative attributes(properties)
2. Gather data and evidence
3. Identify corrective actions to prevent recurrence of harmful effects.
4. Identify solutions
5. Identify useful methodologies for problem solving and problem avoidance.

Structured WHAT IF Technique (SWIFT)


SWIFT is a hazard analysis method that uses a structured brainstorming
with guide words (key words). The idea is to identify risks with the aim of
being quicker than intensive methods like FMEA.

Failure Rate () = is the frecquency with which a system/component


fails, expressed in failures per unit of time.
4

MTBF = mean time between failures

NTF = no fault found due to oxidation, defective connections of electrical


components, temporary shorts or opens in circuit, software bugs, operator
errors.

THE CRITICAL FIVE


1. Direct Cause

= the cause that directly resulted in the event

2. Contributing Cause = the cause that contributed to an event but


itself would not have
caused the event (the cause after
the direct cause)
3. ROOT Cause
= the fundamental reason for an event, which if
corrected would
prevent recurrence.
4. Specific Corrective Action = actions taken to correct/improve
conditions noted in the
event, by changing the direct
cause or the direct cause and
the effect.
5. Preventive Corrective Action = actions takent that prevent
recurrence of the condition
noted in the event
Example:
Keep asking WHY?
Problem : didnt get to work on time Why?
Direct Cause: Car wouldnt start Why?
Contributing Cause: Battery was dead Why?
Contributing Cause: Dome light on all night Why?
ROOT Cause: Kids played in car, left door ajar.(intredeschis)

CORRECTIVE ACTION
1. SPECIFIC Specific corrective actions are actions taken to correct the
direct cause. Does
not prevent recurrence
2. PREVENTIVE Preventive corrective actions are actions taken that
prevent recurrence of
the condition noted in the event. It focuses
on changing the root cause and
any contributin cause(s).
3. SYSTEMATIC Systematic corrective actions are actions taken that
address the failure in
the suppliers quality system that allowed
the event to occur.

EFFECTIVENESS MEASUREMENT
Effectiveness measurement is a criteria used to evaluate if the
corrective actions achieved the desire outcome, result. Periodic checks
are necessary to be sure the corrective actions are still in place and
continue to be effective.

EVENT

1.FORM

2.

TEAM

DESCRIBE
ISSUE

GATHER

DATA

3.

IMPLEMENT,
VERIFY
CORRECTIVE
ACTIONS

DIRECT

4. DEFINE, VERIFY CORRECTIVE ACTIONS


ROOT

CONTRIBUTING

DETERMINE CORRECTIVE ACTIONS

SPECIFIC

PREVENTIVE

5. VERIFY CORRECTIVE ACTION


N
O

6. IMPLEMENT PERMANENT CORRECTIVE ACTIONS


7. PREVENT RECURRENCE
SOLUTION ACCEPTABLE
CONGRATULATIONS TEAM

BASIC CAUSES OF A RISK EVENT


1. Physical material item failed
2. Human people did something wrong/did not do something they were
supposed to.
3. Organizational

YE

v
O
d
R
k
T
V
Y
IW
H
M
F
A
E
B
r
t
e
a
n
lP
s
iy
D
b
h
o
g
m
C
c
u

METHO
DS

-it can be used a combination of 2 RCA methods for better results (ex: 5
WHYs + Fishbone Diagram)
FISHBONE DIAGRAM includes people, methods, machines, materials,
measurements, environment.

QUALITATIVE ASSESSMENT
-developed to rank the risk events based on severity
-informations can be implemented into a heat map.
-used in the first stages of a project

QUANTITATIVE ASSESSMENT
- how much capital is sufficient
-Monte Carlo Simulation probability of certain outcomes by running
multiple trial
runs, called simulations, using
random variables
- efficient frontier shows the best possible expected level of return for
its level of risk.
-used in the later stages when more data is available

BOTTOM UP / TOP DOWN


There are 2 methods in which data can be analysed. Bottom up
method not used anymore because it is too expensive.
For redundant items of equipment carrying out the same duty, if one
item has been analysed to component level it is reasonable to assume
that the other item will behave the same as the first item.
RELIABILITY BLOCK DIAGRAM (RBD) = shows the operational
relationships between each
subsystem or
component
OCCURRENCE RATING:

106

Remote failure is unlikely, fail rate

Very Low isolated failure, fail rate

Low isolated failure with similar systems, fail rate -

Moderate occasional failure, fail rate

High generally associated with systems which often fail, fail


rate

105
104

103

102

Very High failure is almost invitable, fail rate

101

SEVERITY RATING:

Minor doesnt cause real effect on system. Failure might not be


noticed
Severity level is MINOR

Low causes a slight deterioration of system capability


Severity level is MINOR
8

Moderate causes some deterioration in system capability


need for unscheduled work/repair; can cause minor injury to user
Severity level is MARGINAL
High causes loss of system capability; may cause serious health
hazard or
serious injury to user;
Severity level is CRITICAL
Veri High could cause system loss and/or death of user(s);
Severity level is MAJOR

DETECTION RATING

Very High Controls will almost certainely detect the existence of


a defect
High Controls have good chance of detecting the existence of a
defect
Moderate Controls may detect the existence of a defect
Low Controls have a poor chance of detecting the existence of
a defect
Very Low Controls probably will not detect the existence of a
defect
Absolute Certainty of Non-Detection Controls will not / cannot
detect the existence of a defect
CRITICALITY = Measure for gravity of failure mode effects and
is determined from combination of severity and the likelihood of
occurrence.

Basic INFORMATIONS recquired for FMEA


process
What does the system do? MISSION
What is its function? FUNCTION
How could it fail to perform its function? FAILURE MODE
What happens if it fails? EFFECT of failure
What is the likelihood of failure? OCCURRENCE (O)
What is the consequence of failure? SEVERITY (S)
What is the predictability of failure ? DETECTABILITY (D)
What is the Risk Priority Number (RPN)? RPN = O x S x D
Risk Priority Number (RPN)
If RPN>33 then consider the possibility to renew/replace the asset
DAMAGE MODE and Effects Analysis = Is the analysis that
determines the extent of
damage on
the system/equipment

10

You might also like