Di erences Between Condition-Based,

Predictive, and Prescriptive Maintenance

SENSORS & DATA ACQUISITION 2 COMMENTS

SHARE THIS

   

BY STEVE HANLY

SHARE THIS

   

There are a lot of data-driven maintenance strategies, how do you know which one is right for
you and your assets? How do you know if any are even better than simple preventative
maintenance or no proactive strategy at all?!

In this article, I will provide an overview of the following 5 types of maintenance using lots of
graphics: 

1. Reactive Maintenance
2. Preventative Maintenance (PM)
3. Condition-Based Maintenance (CBM) 
 
 4. Predictive Maintenance (PdM) 
Sign Up for Sensing Tips!
5. Prescriptive Maintenance
Thanks for submitting the form.
I'll compare these maintenance strategies on a P-F curve; their return on investment (ROI); which
types of assets they align best with; and how they evolve into each other. The big takeaway from
this post is that in order to know which maintenance strategy works best for you, you need to
start by taking some data!

As you can see, there's a lot to cover in this post!:

1. Types of Maintenance
a. Reactive
b. Preventative (PM)
c. Condition Based Maintenance (CBM)
d. Predictive Maintenance (PdM) 
e. Prescriptive Maintenance
2. Cost and ROI
3. Picking a Maintenance Strategy Based on Machine Asset Type
4. How to Get Started with Prescriptive Maintenance- Exploration!
5. Conclusion and Additional Resources

Types of Maintenance

P-F Curve of Di erent Types of Maintenance

To de ne and understand the di erence between these types of maintenance, let's rst look at a
P-F curve which starts with a potential failure (P) and ends with a functional failure (F).


 
 Sign Up for Sensing Tips!
Thanks for submitting the form.

Reactive maintenance waits for the system to experience a functional failure before
maintenance occurs, so this doesn't even appear on the P-F curve.

Preventative maintenance performs maintenance on some xed schedule ideally aligned to be

slightly shorter than the typical P-F interval for a particular machine. This will ideally let it catch a
system while it is in the middle of the curve. The trouble is that the equipment may deteriorate
too fast for the preventative maintenance schedule to correct it in time. Or, the preventative
maintenance may be wasteful if it is " xing" a perfectly ne piece of equipment.

Condition-based maintenance uses sensors and data with preset conditions or thresholds that
when met will signal maintenance is needed. The trouble is that this requires a fairly noticeable
amount of degradation to have taken place in order to hit these thresholds.

Predictive maintenance uses sensors and data to detect trends in the health of a system and
predict when failure will occur. This allows it to detect the deterioration of a machine earlier than
CBM and allows maintenance teams more time to schedule maintenance at a convenient time,
knowing when the PdM predicted failure to occur.

Prescriptive Maintenance uses sensors, data, and advanced analytics to determine the route
cause of a potential failure so speci c corrective action can be prescribed. The advanced data
and analytics needed for successful prescriptive maintenance also ensures the potential failure
is identi ed even earlier which makes xing the problem easier and less expensive.


 
 Sign Up for Sensing Tips!
Evolution of Maintenance Based
Thanks on Data
for submitting the form.

Another way to look at the di erence between maintenance types is to understand their
dependence on data. This is the most de ning distinction between these maintenance
strategies. 

Reactive maintenance has absolutely no dependence on data, it waits for failure.

Preventative maintenance uses either time or usage to inform its maintenance schedule, it's
not a lot of data - but it's a start!

Condition-based maintenance will typically look at a single metric and check against preset
thresholds.

Predictive maintenance typically looks at a handful of data sets and relies on slightly better
analytics to pick up on various trends in the data and the health of the asset.

Prescriptive maintenance uses many data sets and metrics and likely some proprietary
analysis techniques to determine the root cause of a potential failure.

Now that we've given a good overview of the di erent maintenance strategies, let's dive a bit
deeper into each one.

Reactive Maintenance


 
 Sign Up for Sensing Tips!
Thanks for submitting the form.

Reactive maintenance is just as it sounds: wait until the machine fails and then x it. In general,
this is often not really a maintenance strategy -- it is often the result of having no strategy at
all. Once the machine goes down, the team scrambles trying to nd a x and they feel like they
are constantly putting out res!

There is an exception, though, when you can make reactive maintenance an actual strategy that
has some value. If your machines are:

1. Easily xable
2. Replacement parts are readily available
3. Downtime is not costly, there are redundant machines

Then a reactive maintenance strategy can be a good idea assuming you are planning for this and
it is not costly in both time and money. This is called a "Run to Failure" strategy.

Bene ts and Pros of Reactive Maintenance

No up front cost or cost to manage

Minimal maintenance sta needed

Downsides and Cons of Reactive Maintenance

Unpredictable and inconsistent downtime

High costs associated with xing a failed machine
Repairs tend to be urgent which may require overtime labor

 
 Sign Up for Sensing Tips!
Preventative Maintenance (PM)
Thanks for submitting the form.

Preventive maintenance is using some routine, either time or usage, to de ne when you replace
or rework some components of your system. An example of this is how you replace the oil in
your car every 5,000 miles.

This is a good idea to prevent any deterioration in the machine from occurring in the rst place.
It is easier to plan and work around and it's also generally less expensive to replace a simple
component a few times than repairing a damaged machine after failure. But this has a few
notable issues:

Schedule may be too frequent - you may be paying for maintenance on a perfectly
healthy machine, which is wasteful in labor and parts
Schedule may be too infrequent - your maintenance window may miss a rapidly
accelerating failure
Lulled into a sense of security, failure may arise from another cause - the preventative
maintenance can only address the common failure modes; replacing your oil won't prevent
a belt failure, for example

The more complex your machine is, the more unlikely preventative maintenance will be a
feasible maintenance strategy. This is because it will require too many di erent things to

 
 maintain (all at di erent rates) making it probable that one of those things were excessively xed
Sign Up for Sensing Tips!
(wasteful) or that a failure was missed altogether.
Thanks for submitting the form.

Bene ts and Pros of Preventative Maintenance

Increases asset life span

Reduces unplanned downtime
More e cient use of energy and resources

Drawbacks and Cons of Preventative Maintenance

Schedule may be too frequent (wasteful)

Schedule may be too infrequent (misses a potential failure)
Can't catch uncommon failure modes

Condition-Based Maintenance (CBM)

Condition-based maintenance uses sensors mounted to a machine to periodically check the
health of that asset. Then it compares the most recently measured value against preset
"conditions" to be reached which signify an alarm or warning. Typically this is done by:

1. Establishing a baseline of a known good or healthy machine

2. Calculate a normal distribution of the data taken
3. Set warning levels at +/- 3 standard deviations and then alarms at +/- 5 standard deviations
o the baseline levels

In the example below, about 70 hours into the pro le the velocity RMS (a common vibration
metric to use) on this machine reaches the warning level which a condition-based maintenance
system could detect and report.


 
 Sign Up for Sensing Tips!
Thanks for submitting the form.

Many condition-based maintenance systems require the user to de ne warning and alarm
thresholds. Some "advanced" ones will automatically do that baseline determination for a period
of time and then calculates a simple normal distribution to set warnings & and alarms using the
standard deviation.

Bene ts and Pros of Condition-Based Maintenance

Increases asset life span

Real-time (and often remote) indication of the health of an asset
Maintenance only performed when needed
Advanced warning provided to allow for convenient scheduling of maintenance

Drawbacks and Cons of Condition-Based Maintenance

Requires monitoring hardware and software to generate warnings which have upfront and
recurring costs
Some software may require the user to be specially trained to operate

Predictive Maintenance (PdM)

In the example above about condition-based maintenance, there was clearly a trend in the
data before it reached the warning level. This trend is exactly what a more advanced

 
 algorithm would be able to detect and is more accurately considered "predictive." Predictive
Sign Up for Sensing Tips!
would then go that added step further and predict how much usable time remains in the asset
Thanks for submitting the form.
to schedule maintenance -- it predicts when failure will occur. 

Predictive maintenance (PdM) is not magic; it is about measuring one or many parameters and
using this data to make an informed prediction about how the health of the machine will look in
the near future.

In order to have this data-driven crystal ball, predictive maintenance solutions typically require
more & better data as well as advanced analytics compared to CBM solutions. And this will often
require some machine learning algorithms to be e ective.

Bene ts and Pros of Predictive Maintenance

Increases asset life span

Real-time indication of the health of an asset
Maintenance only performed when needed
Advanced warning provided to allow for convenient scheduling of maintenance

Drawbacks and Cons of Predictive Maintenance


 
 Requires monitoring hardware and software to generate warnings which have upfront and
Sign Up for Sensing Tips!
recurring costs
Thanks for submitting the form.
Some software may require the user to be specially trained to operate

Prescriptive Maintenance
If predictive maintenance uses data to tell you WHEN the machine may fail, then prescriptive
maintenance uses more data to tell you WHAT SPECIFIC PART will fail in the machine. It tells the
maintenance team how to x the problem, not just tell you that there is a problem. How do you
do this? Well... you need a lot of data!

In order to have a prescriptive maintenance program, you need to be able to correlate speci c
signatures in the raw data and trends to speci c root causes. And this tends to require you to
have been able to previously nd these signatures after the fact.

Let's look at an example from the US Navy using enDAQ sensors detailed in this case study. The
US Navy begins seeing a rise in physiological episodes that put crewman and aircraft at risk --
this was both a safety and nancial risk [see news article]. Physiological episodes were
manifesting as pilot hypoxia symptoms thought to be a result of cockpit air contamination, lack
of oxygen, or issues with the Environmental Control System (ECS).

They responded to this issue by:

1. The US Navy needed data to determine the root cause of the physiological episodes and
turned to enDAQ, a convenient, adaptable and ight quali ed data acquisition system
(originally developed for vibration, but also included other sensors such as pressure) to
measure pressure inside the cockpit.
2. The US Navy began measuring the pressure pro le of EVERY F-18 ight of the Navy around
the world -- that is a lot of data!
3. Data was rst used to identify when there was an episode and help treat the pilot -- rst
using condition based maintenance.
4. But then data scientists began looking back in the data in search of indicators that could

  have predicted the event -- now using predictive maintenance
 5. And with even more data, scientists started to be able nd more intricate signatures to tell
Sign Up for Sensing Tips!
maintenance teams which speci c part to x -- evolved into prescriptive maintenance
Thanks for submitting the form.
This is what is pictured in the image above showing the table of metrics (that
correspond) to various components within the environmental control system and
particular operating states during the ight.
Looking at this data (the data has been modi ed from its original form for security
reasons), the Navy is able to make maintenance decisions and inform their crew what
particular parts need to be xed.

To quote the Rear Admiral Fredrick Luchtman:

“...we are almost to the point with our data analytics where we receive information
from the eet, we analyze that information, and then we can tell the eet, hey this
particular aircraft is exhibiting signs that this particular part may be needing to be
replaced pretty soon.”

This quote and example is especially import to highlight that prescriptive maintenance (and
predictive for that manner) requires an evolution: You start by being in a state where you have a
problem and are reacting to failures. Then you begin doing some preventative maintenance to
and in that may start periodically gathering data. This data can then start being used as a
condition-monitoring program. With more time and more data, you may have enough to start
being able to use that to predict failure. And with even more data you begin being able to nd
root causes and prescribe the speci c corrective action ahead of time. It's an evolution, so let's
revisit that earlier image!


 
 Cost and Return on Investment
Sign (ROI)
Up for Sensing Tips!
Thanks for
When looking at the cost of a maintenance submitting
strategy, the form.
there will be a few considerations:

Labor
Recurring: Time spent by maintenance personnel to execute the strategy on a daily
basis
One-Time: Training to understand the maintenance equipment & software (if any)
Monitoring Equipment & Software (if any)
Recurring: Software charges to license the analytics and alerting capabilities
One-Time: Purchasing and installing the sensors used as the data source

These maintenance investments will (hopefully) have a return on said investment in the form of:

Reduced cost of repairs: This is because preventative and predictive maintenance

strategies are generally xing something "small" before it turns into a catastrophic and
expensive failure.
Increased Asset Uptime: Machines are running only if they are generating a pro t which
means if a machine is down due to a failure you have an opportunity cost because it isn't
making money.

The plot below compares our maintenance strategies and illustrates how they stack up on that
return on investment.

Reactive Maintenance
No cost to install and manage because you're doing nothing!

 
 But it also is going to have the highest cost of repairs because it requires xing
Sign Up for Sensing Tips!
catastrophically failed machines and doing so urgently. And this strategy is likely to
Thanks for submitting the form.
have machines down for longer periods of time. But as we'll see in the next section,
there are some times this still makes sense.
Preventative Maintenance
Linear/proportional increase in both cost & savings
It has a big jump up in costs to manage because you will now be paying people to
perform maintenance on a schedule and paying for these parts and material needed
in the smaller maintenance tasks.
But this comes with a proportional increase in savings from both cost of repairs and
increased machine uptime. The preventative maintenance will without a doubt
reduce the likelihood of failure and downtime while also generally having cheaper
repair costs as it replaces inexpensive components.
Condition-Based Maintenance
Slightly higher increase in savings relative to costs
Condition-based maintenance requires some upfront investment in the monitoring
equipment and training. It also generally will have some recurring cost associated
with software licenses and paying people to pay attention to the data and alarms
(and some require humans to manually capture the data).
But you will have a slightly higher proportional increase in savings as you ensure
repair costs are only allocated to machines in need of it. And this will provide greater
peace of mind (rightfully so) that you won't have unexpected downtime of a machine
providing greater savings.
Predictive Maintenance
Higher increase in savings relative to cost increase
If you already have a condition-based maintenance strategy then the only increase in
costs will be for better & more data, and better analytics, which is relatively small to
the initial investment for CBM in the rst place.
But predictive maintenance will further increase machine uptime as failures can be
predicted and maintenance scheduled conveniently. Costs will also generally be less
because the failure was caught even earlier.
Prescriptive Maintenance
Much higher increase in savings relative to cost increase
Like the di erence between CBM and PdM, the step up in costs from preventative to
prescriptive is also relatively small - again only for better & more data with better
analytics. But the most important thing it needs relative to predictive is just even
MORE data and history, which isn't all that expensive.
Prescriptive maintenance o ers a massive increase in savings however as you are
able to tell maintenance teams exactly what to x (and what parts to bring with them)
without needing them to do onsite evaluations. This also will provide even further
improved machine uptime as failures are detected even earlier.


 
 Picking a Maintenance
Sign Up Strategy
for Sensing BasedTips!on Machine
Asset Type Thanks for submitting the form.

There isn't a one-size- ts-all maintenance strategy. The best strategy depends on the machine
asset's complexity and criticality. The chart below breaks these two variables into 4 quadrants
and plots where each maintenance strategy aligns best for that type of machine.

Simpler and Not-Very Critical Assets may get away with a reactive strategy depending on
how easy they are to repair and how inexpensive downtime costs. But as either the
complexity to repair or the criticality of the asset increases, a preventative maintenance
strategy will be the most cost e ective.
Simpler Yet Critical Assets will bene t most from a condition-based maintenance
strategy to prevent downtime while being e ective at detecting failure of the simpler
machine. Prime examples of these machine types are bearings that operate at generally
one speed and pumps.
Complex & Critical Assets will bene t most from predictive or prescriptive maintenance.
More complex machines require more than just one variable to monitor and the ability to
understand di erent operating states and what is "good" and "bad" signatures in those
varying states. And the improved analytics will enable even less downtime of your critical
asset to keep it running and generating money!
Complex Yet Not-Very Critical Assets don't really have a good t. It depends on just how
expensive it is to repair a catastrophic failure. It is unlikely that CBM will work for these
type of assets, though, because operating conditions will require many variables to be
monitored and analyzed. High repair costs may warrant predictive or prescriptive 
 
 maintenance, but low repair costs may allow some preventive maintenance, or even
Sign Up for Sensing Tips!
reactive maintenance, to make sense.
Thanks for submitting the form.

How to Get Started with Prescriptive Maintenance -

Explore!

Hopefully the bene ts of prescriptive maintenance are intriguing, but how do you get started!?
Well, you can't start o by doing prescriptive maintenance -- you can't even start o by doing
predictive maintenance. The way you get started is by getting some data on your machine at
a few di erent times.

Ideally, you are already doing some preventative maintenance and so you are utilizing a
schedule to go out to the asset, inspect it, and potentially repair parts. Use these opportunities
to gather some vibration, temperature, and other sensor data on your asset.

You don't have to use a fancy wireless monitoring system for this, either. You can use a simple
vibration meter or data logger (See our blog post 6 Ways to Measure Vibration for an explanation
of the full range of vibration measurement systems readily available to a test engineer).

As you go through the process of collecting data on your asset, you will be able to determine: 
 
 How accessible is your asset for sensors?
Sign Up for Sensing Tips!
What type of vibration signature is present in the asset? Does it look erratic or steady from
Thanks for submitting the form.
one test to the next?
What metrics seem to show some valuable insight? Do you need to be looking in the
frequency spectrum or just simply the velocity RMS?

Once you're able to answer these questions, you'll be able to understand if a data-driven
maintenance strategy can work for you and your machines. Then you can begin expanding your
"exploration" period. This data-driven maintenance exploration will eventually evolve into a
prescriptive maintenance strategy over the course of time with continuous data
collection and data analysis. 

Like I demonstrated earlier with the maintenance evolution chart, prescriptive maintenance
requires an evolution. All you have to do to start the process is start taking some data!

Conclusion and Additional Resources

I hope that after reading this post that you now have a good understanding of the di erences
between reactive, preventative, condition-based, predictive, and prescriptive maintenance and
that you're ready to start collecting some data and exploring how you can perform prescriptive
maintenance on your own assets. 

If you have any questions, please don't hesitate to leave a comment or contact us directly. We
are here to help you with all your testing and analysis needs! And, if you liked this post, I hope
you'll subscribe to our blog for monthly updates on monitoring, vibration analysis, data
acquisition and more!

Related Posts:

Top 9 Accurate Wireless Vibration Monitoring Systems

Top 12 Vibration Metrics to Monitor & How to Calculate Them
Why the Power Spectral Density (PSD) is the Gold Standard of Vibration Analysis


 
 Sign UpSHAREfor Sensing
THIS Tips!
   
Thanks for submitting the form.

Steve Hanly
Steve is the Vice President of Product at Mide. He started out at Mide as a Mechanical
Engineer in 2010. He enjoys getting his hands dirty to do some...

W H AT AR E YOUR T HOUGHTS?

Comment
Enter your comment here...

First Name

Last Name

Email


 
 protected by reCAPTCHA Sign Up for Sensing Tips!
Privacy - Terms
Thanks for submitting the form.

Post Comment

By signing up you are agreeing to our Privacy Policy

COMMENTS (2)

Gustavo Antonio 11/25/2020, 4:40:10 AM

me gustaría si me pueden enviar el archivo de FFT, PSD, and Spectrogram

Reply to Gustavo Antonio

Gustavo Antonio 11/25/2020, 4:40:31 AM

me gustaría si me pueden enviar el archivo de FFT, PSD, and Spectrogram

Reply to Gustavo Antonio

   

