CMRP Training

The Plant and Equipment Wellness Way
to Enterprise Asset Management Success and

World Class Operational Excellence
3-day training course

DAY 1
www.lifetime-reliability.com 1
PEW/PWW Course Content
Day 1 Day 2 Day 3

Foundations PWW Processes Reliability Creation
• Physics of Failure • Risk Identification • Business Risk Reduction
• Reliability • Risk Selection • Stress to Process Model
• Risk • Risk Control Planning • Life Cycle Risk Reduction
• Cost of Failure • Risk Control Introduction • Operational Risk Reduction
• Series Arrangements • Risk Monitoring • Machinery Risk Reduction
• Human Error • Risk Continual Elimination • Making Changes
• Life Cycle
• Reliability Improvement
Why do Machines and Equipment
Continue to Fail in Companies?
• “We get reliability by creating and building a thing that can do the
duty, and preventing its failure during use.” (LRS uses Plant Wellness Way to do that.)
Day 1
Foundations
• Physics of Failure
• Reliability
• Risk
• Cost of Failure
• Series Arrangements
• Human Error
• Life Cycle
Machines Fail because their Parts First Fail
2nd bearing sleeve 2nd bearing bush
1st bearing sleeve 1st bearing bush
Understand How Machines are Designed
TIP: THE SECRET TO GREAT EQUIPMENT LIFE IS TO …
KEEP PARTS WITHIN THEIR DESIGN STRESS ENVELOPE!
L3
Size of a L4 L2
human
L1
hair
25 - 0.025
+ 0.025
- 0.01
25 + 0.01
Ted, when they design machines, like this shaft rotating in two bearings, they keep the parts in place by
making the gaps between them very small. The hair on your head is about 0.1 mm (0.004”) thick. On this
25 mm (1”) shaft, the gap between the metal surfaces can be as small as 0.01 mm (less then 0.0005”).
That is 10 times thinner than the thickness of your hair. That is very little space for things to move in. If
the parts get twisted and distorted then that clearance disappears and you have parts hitting each other.
Any machine in that situation will quickly fail.
The Unforgiving Nature of Machine Design
How far off-center did the designer allow the shaft to move?
How much movement/angle did the bearing designer allow?
How much distortion before the parts overload and fail?
The parts’ engineering clearances mean that everything has to be exactly as the designer
planned it to be. The whole machine needs to run precisely as it should. If parts are deformed
outside of their tolerance, like in this sketch, then the bearings will fail in a matter of hours, and
not the years that they should last in a machine that is working as it was designed to operate.
Remember: The Limit of Machine Distortion is set by Design Tolerances – don’t let a
machine or its parts get twisted out of shape!
Stress from Distortion
Point contact only

Cantilever causes
distortion when bolted
down
Shaft misalignment distorts and

bends shafts which in turn overloads
the shaft bearings
Source: Shaft Alignment Handbook, John Piotrowski, CRC Press
Far too common examples of
soft-foot problems!
The Physics of Parts Failure
Range of Factor of Safety
Range of Forces cause stress in the part when a load
Operating
Material contacts the part. The smaller the contact area
Stress OVERLOAD causes the greater the induced material stresses. We
local stress to rise Strength show the pattern of varying operating loads that
Frequency
a part can experience at the contact points as a

curve from least load to most load.
Parts with least strength

fail when overloaded
Parts fail whose strength

Size of Stress
is weaken to this level Material strength falls
from FATIGUE
Frequency
Parts ‘age’ as they are used. Loads stress the

physical structure and it breaks under high
loads. The weakest parts fail early; the
strongest take more stress before they too fail.
We show the degradation as a curve of material
strength from most strong to least strong.
Size of Stress
Why do parts fail? Because they can no longer handle the stress they suffer. When the load is too
great the part fails from ‘overload’, when the material weakens and degrades it fails from ‘fatigue’.
Fatigue Limit of Parts’ Material of Construction
Failure
Failure
10,000 cycles Limited life at Infinite cycles

1,000,000 cycles
at this stress this stress level at this stress
at this stress level
level for non-ferrous level for steel
We must know what our equipment parts are made of and prevent high stress in those
with infinite life but replace those of finite life before they fail.
The Equipment Designer Wanted a Long,
Trouble-Free Service Life
Strength
Failure
Factor of
Safety
Failure
Load
Using a Factor of Safety means the Designer

intended for the operating stress to be so low
that it produced a long, trouble-free service life
Stress at Atomic and Microstructure Levels Destroy
Uncoil a paper clip and 1) bend it and 2) twist it as instructed by the
Presenter. Carefully count the number of cycles until it breaks.
No of Stress Cycles Before Failing

Proportion of Ultimate Stress for
Failure
For long operating life,
parts need to stay
Wrought Steel
Failure below the infinite life

atomic stress level
20 cycles at this 200,000 cycles at Infinite cycles at

stress level this stress level this stress level
Have you ever bent a metal wire back and forth until it breaks from being worked? If you have then you were performing
a stress life-cycle test. The wire does not last long when severely bent one way and then back the other way. Each bend is
an overstress, and eventually the overstressing accumulates as damage to the microstructure. The wire fatigues and fails.
The very same thing happens with the parts in yourwww.lifetime-reliability.com
machines. If you want your parts to NEVER FAIL FROM OVERSTRESS 12 –
KEEP OPERATING STRESSES BELOW INFINITE LIFE LEVELS.
Activity – Stressing Steel Parts
Paper Clip Failure Distribution Curve
Uncoil a paper clip and 1) bend it and 2) twist it as instructed by the
Presenter. Carefully count the number of cycles until it breaks.
Develop a distribution of the count of the number of
1) bending cycles to fatigue, and
2) torsion cycles to failure.
14
13
12
No of Failures
11
10
9
8
7
6
5
4
3
2
1
0
1 2 3 4 5 6 7 8 9 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 6
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0
No of Cycles
Impacts of Out-of-Roundness
Spalling
What it should be..
1500RPM = 25
impacts/second
3000RPM = 50
impacts/second
Failing Roller Bearing Degradation Curve and
the Worsening Defect Severity
Stage 1.
Approx 10% to 20% Stage 2.
remaining life 5% to 10%
Failure Ultrasonic remaining life Stage 3.
Induced Energy Vibration Analysis Oil Analysis 1% to 5%
Detected Fault Detection Detected of remaining life
Negligible P. Low Risk Temperature Rise

Risk
F.
Part Condition
Audible Noise
Stage 4.
To Hot to Touch Remaining life
one hour to 1%
Mechanically
Loose
Ancillary
Damage
PRECISION
OPERATION AND
OPERATOR CARE MAINTENANCE
PREDICTIVE
PREVENTIVE (need to consider the
RUN TO FAILURE
probability of detection)
Catastrophic
Failure
Time
www.lifetime-reliability.com Source: Ricky Smith, Allied Reliability, 2009 15
Machinery Lubrication Article (5/2007), with additions
Operating Stresses Cause Failure
Source: Extract from ‘Mobile Plant Maintenance and the Duty Meter
Concept’, Hal Gurgenci, Zhihqiang Guan, Journal of Quality in Maintenance
Engineering, Vol 7, No4, 2001.
Walking Dragline
Production 30m 50m 28m
Tip: Because each operator handles the dragline differently, at their own work rate, there are varying stresses placed
on it. The cumulative wear on the machine is not consistent hour after hour, so using an hour-based preventive
maintenance period is inappropriate; you may be maintaining too early, or too late. The right way is to also count the
stress peaks and estimate how much life each one destroys and add that to the usage meter.
The Operating Overload Cycle
Many parts fail

without exhibiting
warning signs of a
Smooth Running
coming failure – they
Operating Performance
show no evidence of An Overload
Smooth Running
degradation; there is Another Smooth Running

just sudden Overload Potential operating life
The ‘Death’ lost; now curtailed and
catastrophic failure. Overload wasted
In such cases the
parts were too weak
for the loads they
had to take. In
virtually every case
Failed!?
those loads are Now you have
imposed by human Time (Depending on the situation
to investigate!
error. this can be at anytime.)
The Stress-Driven Failure Degradation Sequence
Cause of Aging Failures
Time Dependent Load and Strength Variation The strength
distribution widens
Strength and falls over time.
An Another
The ‘Death’
Overload Overload
Overload
Likelihood of
failure is higher in
Load this region
Equipment replaced here – Few Problems! Time/Load Cycles

Log Scale
Equipment replaced here – Lots of Problems!
Estimated Life
Probable Life Uncertainty Wear-out
Zone
Rate that
parts fail
Time
Building for the Physics of Failure
Design in Reliability and Low Operating Risk
Operating/Maintenance Cost Management
Failure
Mode
Environment
Effects
and
Criticality
Operating
Analysis
Stresses
Life Cycle
Management
Strength Reliability
Of the Engineering
Material
Source: Pecht, Michael., ‘Why the traditional reliability prediction models do not work - is there an alternative?’,
CALCE Electronic Product and Systems Center of the University of Maryland, College Park, MD, 20742, USA.
PEW SOLUTION: Physics of Failure Causes of
Atomic and Microstructure Stress
Day 1
Foundations
• Reliability
• Risk
• Cost of Failure
• Human Error
• Life Cycle
What is Reliability?
• “Reliability is the probability that an item of plant will perform its
duty without failure over a designated time.” (Formal Definition)
• “Reliability is the chance of completing the mission.” (Military Definition)
• “Reliability is the chance of success.” (LRS Definition)
• “We get reliability by creating and building a thing that can do the
duty, and preventing its failure during use.” (LRS uses Plant Wellness Way to do that.)
Machine Reliability = Sum of Parts’ Reliability
Failure from Error Failure from Induced Stress Failure from Usage
•Defective parts •Operating overload •Too many aging parts
•Poor quality assembly •Rapid aging of some parts •Many parts degraded
•Manufacture error •Local environment degradation
System •Operator error Mean of Many
The ‘failure Rate of 50-70% •Poor operating practices 10-30%
curve’ for a Systems
Failing
•Poor maintenance practices (many machines)
machine has
•Poor design choice 20-30%
a special
name
– ROCOF –
Rate of A Single System
Occurrence (a machine)
of Failure.
Time or Usage Age of System
Component
Rates of
Failing
Time or Usage Age of Parts
Parts put together into machines form a system of parts. When a working part fails the machine fails.
Hence the reliability of a machine is less than the reliability of its worst part. The ROCOF curve for a machine
reflects what happens to its parts, and moves up and down as parts fail. But when we take many identical
machines and collect their parts’ failure history together, we get a ‘steady average’ ROCOF, which is
representative of the reliability of the www.lifetime-reliability.com
machine design, and its use and care over its lifetime. 23
What is the Chance of this Drinking Glass Breaking?
Its Reliability is, ‘The chance it will hold water next time you use it’
What can cause this glass to break?
• It can be dropped, for example -

Stay with me, because 1. slip from your hand
understanding how to 2. fall off a tray
3. slip out of a bag or carry box
measure reliability is one • It can be knocked,
of the most important 1. hit by another glass
concepts that you need to 2. clanked when stacked on each other
3. hit by an object, like a plate or bottle
know if you want to do
• It can be crushed,
maintenance well. 1. jammed hard between two objects
2. stepped-on
3. squashed under a too heavy object
• It can be temperature shocked,
The many ways for the 1. in the dish washer
2. during washing-up
glass to break (the failure
• Mistreated,
mode), are called ‘failure 1. It can be thrown in anger
mechanisms’. 2. It can be smashed intentionally
• Latent damage
1. scratched and weakened to later fail more easily
2. chipped and weakened to later fail more easily
Chance of Failure for a Drinking Glass
1,000,000 glasses sold in packs of 12
1 83,333 households buy a pack of 12
Say average household breaks 2 glasses a year • It can be dropped, for example -
1. slip from your hand
That is 166,667 glasses broken each year which are then replaced 2. fall off a tray
Failure Rate per Year

Chance of breaking a glass during a year is 166, 667 ÷ 1,000,000 • It can be knocked,
+
1. hit by another glass
+ 2. clanked when stacked on each other
Chance of Glass Failure Curve + 3. hit by an object, like a plate or bottle
+ • It can be crushed,
+ 1. jammed hard between two objects
0.167 + 2. stepped-on
3. squashed under a too heavy object
+ Crushed - squashed • It can be temperature shocked,
+ Crushed - jammed 1. in the dish washer
2. during washing-up
+ Mistreated - smashed • Mistreated,
1. It can be thrown in anger
+ Dropped - tray 2. It can be smashed intentionally
+ Knocked - stacked • Latent damage
1. scratched and weakened to later fail more
+ Knocked - hit easily
Dropped - hand 2. chipped and weakened to later fail more
0 easily
0 12 24 Time (months)
‘Opportunity’ for
breakage arises
regularly
Stop Breakage = Remove Failure Causes =
Improved Reliability
Design Change
1
• It can be dropped, for example -
1. slip from your hand
2. fall off a tray
Failure Rate per Year

• It can be knocked,
Procedure Change 1. hit by another glass
2. clanked when stacked on each other
3. hit by an object, like a plate or bottle
×
Instructions • It can be crushed,
1. jammed hard between two objects
& Training 2. stepped-on
0.167 3. squashed under a too heavy object
• It can be temperature shocked,
1. in the dish washer
0.045
$ $ $ $ 2. during washing-up
• Mistreated,
1.
2.
It can be thrown in anger
It can be smashed intentionally
+ Mistreated - smashed • Latent damage
+ Knocked - hit 1. scratched and weakened to later fail more
easily
Dropped - hand 2. chipped and weakened to later fail more
0 easily
0 12 24 Time (months)
‘Opportunity’ for
breakage arises
regularly
Reliability = Remove the Chance of Failure
Dropped
Hit/Impact
Total System
10 Yrs
Wear
Puncture
Total System
60,000 km
Misaligned
Insufficient Lube
Wrong Lube
Particulate/Dirt
Moisture
Poor Fit
Overload
Total System
www.lifetime-reliability.com
5 Yrs 27
The Unreliability of Systems of Parts (i.e. a
Machine) is the Sum of Its Parts Failure Rates
The shape and position of the ‘system’ curve is adjustable by varying
the policies controlling quality and maintenance!
The reliability of a
machine is always Quality Replace
Control, Equipment,
less than its parts. System
Training, Add more
Rate of
When one part Failing Precision components to
fails the whole Assembly PM, PdM (Condition Monitoring), Precision Operation PM
machine fails.
With many parts A Single System (machine) Mean of Many
in a machine, Systems (machines)
Component
there are many Rate of
Failing
chances of failure.
Time – Age of System
The Maintenance Zones of Equipment Life
To improve the reliability of a series of parts (that’s a machine) we must improve

the reliability of each part. We must ensure each part gets its maximum life.
“Equipment reliability is malleable by choice of
policy and the quality of practice.”
ERROR INDUCED ZONE STRESS INDUCED ZONE USAGE INDUCED ZONE
• Better quality control • Condition Monitoring • More parts on renewal PM
• Higher skills training • Better operator training • Better material choices
• Precision assembly • Total Productive Maintenance • Considerate operation
• Precision installation • Precision Maintenance • Degradation Management
• No substandard material • Better design/application choice • Timely maintenance
• No manufacturing errors • Stronger material choices
• Machine protection devices
• Robust packaging
• Operator ITLC
• Deformation Management
Old Machine
• Defect Elimination
System • Manage ‘Acts of God’ Better Machine
Rate of
Failing
Time or Usage Age of System

Component
Rates of
Failing
Remove Causes Time or Usage Age of Parts

When we remove
parts’ failure by of Parts’ Failure
changing our policies
and using better
practices, equipment
becomes more reliable www.lifetime-reliability.com ITLC: Inspect, Tighten, Lubricate, Clean 29
Equipment Life vs. Chance of Failure
‘Precise’
‘Smooth’
‘Tight’
‘Dry’
‘Clean’
‘Cool’
‘Repeatable’
Source: Wayne Bissett, OneSteel Reliability Manager, Planning and

Condition Management Presentation, Sydney, Australia, 2008
Where does Failure Start?
Problems start with ‘chance’ variation in our decisions… for example
±1 Std Devn Accuracy Relative
for ‘Feel’ Cost
Number of Events
Feel – Operator judgement ± 35% 1
Torque Wrench ± 25% 1.5
Turn-of-the-Nut ± 15% 3
Load Indicating Washer ± 10% 3.5
Fastener Elongation ± 3 - 5% 15
Strain Gauges ± 1% 20
- 15% + 15% Torque

- 35% + 35%
100% Required Torque
Ajax Fastener
Variation in Torque on a Bolt
Handbook 1999
31
How Chance Tricks Us so We think ‘Feel’ is Fine
±3 Std Devn – 99.7%
±2 Std Devn – 95.5% Accuracy Relative
±1 Std Devn – 68.3% Cost
Feel – Operator judgement ± 35% 1
Torque Wrench
Number of Events
± 25% 1.5
Turn-of-the-Nut ± 15% 3
Load Indicating Washer ± 10% 3.5
Fastener Elongation ± 3 - 5% 15
Strain Gauges ± 1% 20
- 15% + 15% Torque

Opportunity!!
Variation in Torque on a Bolt
- 35% + 35% Opportunity!!
Opportunity!!
100% Required Torque
Opportunity!!
Ajax Fastener Handbook 1999
We can imagine situations that will cause parts failure
Identify Equipment Assemblies and
Parts at Risk of Failure
*^ * Wear-out (age/usage related failure) > PM inspection
^ + From Usage (contaminate with use) > PM renewal
^
^
• Induced Stress (random failure) > PdM condition
^ > PrM/PrO precision
^
^
^*
^ ^ Installation Error (early life failure)> PrM/PrO precision
^* > ACE 3T procedures
^
^
^*•
^*
^
^+
^
** •
* •^ www.lifetime-reliability.com 34
^+
Day 1
Foundations
• Reliability
• Risk
• Cost of Failure
• Human Error
• Life Cycle
Activity – What is the Chance of Failure?
Chance of Failure = 1 – Chance of Success
Chance of Failure = 1 – Reliability
Roll Number No. of Times 1 Appears in 5 Dice
1 0
2 2
3 1
4 2
5 1
6 1
7 0
8 1
9 0
10 0
50 Dice 8
In a large number of random historic events the average chance does not accurately represent
the individual event chance. YOU CANNOT PREDICT THE OUTCOME OF A FUTURE RANDOM
EVENT BY USING PAST RESULTS (EXCEPT BY LUCK).
Estimated Life
Probable Life Uncertainty Wear-out
Zone
Rate that
parts fail
Time
PEW SOLUTION: Reduce the Chance of Failure
Chance of Failure = 1 – Chance of Success = 1 – Reliability
Risk = Consequence $ x Likelihood /yr

Risk = Consequence $ x [Freq of Opportunity /yr x Chance of Failure at Each Opportunity]
Risk = Consequence $ x [Freq of Opportunity /yr x {1 – Reliability at Each Opportunity}]
Now we have Time/Usage related Implications Now we have Time/Usage related Implications
Must Stop Deformation
Only Excellent Only the Correct Only Proper Fits Absolutely No

Lubricant Fastener Torque and Tolerance Unbalance
Cleanliness
Here are some opportunities…
The Risks You Live With and those You
Prevent Show Your Risk Boundary
$1,000K $10,000K Risk = Consequence $ x Chance /yr
If each failure costs your
business $7,000 – $15,000 $100K $1,000K Risk = $1M x 0.01 /yr = $10K x 1 /yr
for every $1,000 of repair $10K $100K
Never Accept
cost … what risk is the $1K $10K
business willing to carry? Accept
$0.1K $1K
How often will a failure Repair Business 0% 50% 100%

event be accepted? Cost per Cost per Chance Of Failure
Event Event in Time Period
• What failures don’t you bother repairing, but immediately replace with new?
(The risks of using rebuilt equipment are too much.)
• Which production equipment will you let fail? (The cost of failure is insignificant.)
• Which production equipment will you never allow to fail? (The cost of failure is too expensive.)
• When will you be willing to replace equipment that you will not allow fail?
(How much remaining life are you willing to give up to reduce the risk of failure?)
• What size safety and environmental failures will you allow? (Their cost is insignificant.)
Acceptable Equipment Failure Domain
Risk = Consequence x [Frequency of Opportunity x Chance of Failure at Each Opportunity]
Repair Business Total
Cost per Cost per What is your tolerance for problems
Failure Event Failure Event on a piece of equipment?
$1,000K $10,000K
$100K $1,000K Outside the Volume Never Accept Failure
$100K Limit of
$10K
$10,000/Yr
$1K $10K
$0.1K $1K
Inside this Volume Accept Failure
10 10% 50% 100%
2 Chance of Failure
1
0.5
0.1
Risk can be Calculated and Plotted
The ‘A’ curve is the same risk throughout
Frequency No/yr
Risk $/yr = Consequence $ x Frequency of Failure /yr

A = Consequence $ x [Opportunities for Failure/yr x Change of Failure]
Too many small failures is just A

as bad as a catastrophe
Consequence $
Risk using Log10 Chance and Consequence
Risk = Consequence x Frequency

Log of Frequency
No/yr
Log Risk =
10 Log Consequence + Log Frequency
0.1
0.01
0.001
1 10 100 1,000 10,000 100,000 Log of Consequence $000

What a Log-Log Risk Scale Means
I used to wonder why we
were so lucky that more
Log Frequency No/yr

things didn’t go wrong! Risk log-log plot
In reality, extreme risk

doesn't arise often.
What is What is
All
the the
threat
likely chance
barriers
cause of the
in place
the ‘holes’
can
‘holes’ line-up
have
in the at the
‘holes’
barriers same
in them.
? time? Log Consequence $
Consequences
Hazard
Want ALARP – As Low As Reasonably Practicable
ALARP Triangle
Intolerable
Risk Rising
Maximum Tolerable Risk
ALARP
Broadly Acceptable Risk
Negligible / Acceptable Risk
High RISK MATRIX

101 10 in 1 yr
A
100 1 in 1 yr
Frequency of Occurrence
10-1 1 in 10 yr
C
10-2 1 in 100 yr
D
10-3 1 in 1,000 yr
E
10-4 1 in 10,000 yr
Low $100 $1,000 $10,000 $100,000 $1,000,000
Lowwww.lifetime-reliability.com
COST High 43
PEW SOLUTION: Asset Engineering, Operations and
Maintenance that Reduces Life Cycle Operating Risk
Engineering, Ops and
Maintenance Required
Actual Engineering, Ops and

Maintenance Performed
Wasted Effort and Wrong Focus
REQUIRED
ACTUALLY PERFORMED
Inadequate Effort and Focus
Equipment Failure Rate
50-70% 10-30% REQUIRED

(ROCOF)
ACTUALLY PERFORMED
20-30% Correctly Matched
Focus with Least Effort
Time or use
PEW SOLUTION: Use a Process to Create Reliability by
Reducing the Chance of Machine Component Failure
Stress Removal Business Wide ACE 3T Lifetime

FMEA/RGCA DAFT Costs Risk Reduction
Life Cycle Operating Risk

Reduction Strategies
MAINTENANCE OPERATIONS ENGINEERING

• Planned Preventive Maintenance • Operate within design envelope • Specifications for reliability –
• Planned Condition Monitoring • Precision Operation stress removal manufacturing, materials, installation,
• Planned Reliability Improvements • Operating Performance Monitoring commissioning
• Precision Maintenance skills and • Operator ‘listen, look, feel’ • Select for life-cycle profit maximising
equipment monitoring and report problems • Design-in reliability, maintainability
• Precision Breakdown Repair • Operator ‘tighten, lubricate, clean’ • Standardise best practices
• Standardise best practices • Standardise best practices
Reliability Growth
Identifying Risks on a Standard Risk Matrix
This layout is the basic approach. There is full mathematical modelling as well, but this basic
table is fine to start with. The layout is universal. You calibrate it by changing consequence
descriptions to what you are willing to accept, and loss costs to what you are willing to pay.
PEW SOLUTION: Uses a Tracking Risk Matrix to
Prove Asset Operating Risk Reduction
DAFT Cost
$1,000,000,000
per Event
$100,000,000
$300,000,000
$10,000,000
$30,000,000
$1,000,000
$3,000,000
$100,000
$300,000
$10,000
$30,000
$1,000
$3,000
Likelihood of Equipment
$100
$300
$30
Failure Event per Year
Event
Descriptor
Count / Time Scale Historic Description 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9
Scale
Year
Twice per
100 2 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5 11
week
Once per
30 1.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5
fortnight
Once per
10 Certain 1 2.5 3 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
month
Once per
3 0.5 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5
quarter
Almost Event will occur on an
1 Once per year 0 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5 9
Certain annual basis
Event has occurred
Once every 3
0.3 several times or more in -0.5 3.5 4 4.5 5 5.5 6 6.5 7 7.5 8 8.5
years
Likely a lifetime career
0.1
Once per 10
years
Possible
Event might occur once
in a lifetime career
-1 3.5 4 4.5 5 5.5 6 CM 6.5 7 7.5 8
Event does occur CM oil condition analysis

Once per 30
$
PM
0.03 Unlikely somewhere from time to -1.5 3.5 4 4.5 5 5.5 6 6.5 7 7.5
years
time CM cable thermographs
Once per 100 Heard of something like
0.01 Rare -2 3.5 4 4.5 5 5.5 6 6.5 7
years it occurring elsewhere
Once every PM oil filtration
0.003 -2.5 3.5 4 4.5 5 5.5 6 6.5
300 years PM oil change
Once every Never heard of this
0.001
1,000 years
Very Rare
happening
-3
PM oil leaks from TX 3.5 4 4.5 5 5.5 6
Once every
0.0003
3,000 years
-3.5 PM water ingress paths 3.5 4 4.5 5 5.5
Once every Almost

Theoretically possible PM oil breather contamination
0.0001 but not expected to -4 3.5 4 4.5 5
10,000 years Incredible
occur PM cable connections
Note: Risk Level 1) Risk Boundary 'LOW' Level is set at total of $10,000/year
Red = Extreme 2) Based on HB436:2004-Risk Management
Amber = High 3) Identify 'Black Swan' events as B-S (A 'Black Swan' event is one that people say 'will not happen' because it has not yet happened)
Yellow = Medium 4) DAFT Cost (Defect and Failure True Cost) is the total business-wide cost from the event
Green = Low
Blue = Accepted
Day 1
Foundations
• Reliability
• Risk
• Cost of Failure
• Human Error
• Life Cycle
The Purpose of Business
I want to
$ show you the
financial
Revenue disaster that
EBITDA Profit plant and
Total Cost equipment
failures cause
a business.
Fixed Cost
Variable Cost
Output / Time
Normal Business Operations
Profit ($) = Total Costs ($) =

Revenue ($) - Total Costs ($) Fixed Costs ($) + Variable Costs ($)
EBITDA = Earnings before Interest, Tax, Depreciation, Amortization – it represents the operating profit.
Impact of Defects and Failures on Profits
Once the equipment fails, new costs and losses start appearing.
Profits
forever lost
Added Cost Impact of a Failure Incident
$
Increased and Wasted Variable Costs

Revenue
Total Cost
Fixed Cost
Wasted Fixed Costs
Variable Cost
t1 Stock-out t2 Output / Time

Effects on Costs and Profit of a Failure Incident
Total Costs ($) = Cost of Loss ($/Yr) =

Productive Fixed Costs ($) + Frequency of Loss Occurrence (/Yr) x
Productive Variable Costs ($) + Costs of Loss ($) Cost of Loss Occurrence ($)
And clearly, repeated plant and equipment failures and
stoppages totally destroy the profitability of an operation.
$ Accumulated Wasted Variable, Fixed and

Failure Costs Revenue Profits
forever lost
Total Cost
Fixed Cost
Wasted Fixed Costs
Variable Cost
t1 t2 t3 t4 t5 t6 Output / Time
Effects on Profitability of Repeated Failure Incidents
If there are lots of failures, you end up running around like headless chooks, losing
money faster and faster. It makes me laugh when I see this happening in a company.
Everyone is busy, but there little profit, … it is all lost in the ‘failure cost surges’.
Benefits of Reducing Operating Risk
$ Accumulated Wasted Variable Fewer profits
Revenue lost, but ‘fire-
and Failure Costs
fighting’ is
high
Risk ($/yr) =
Total Cost
Frequency (/yr) x
Fixed Cost
Wasted Fixed Costs
Variable Cost
Consequence ($)
Effects on Profitability of Reducing Consequence Only
Fewer
$ Profits
Fortunately, we can do something Revenue
Lost
about it. There are two choices – 1)
get very good at fixing failures fast, or Total Cost
2) don’t have failures in the first place
- ZERO DEFECTS is the way to go. Fixed Cost
Wasted Fixed Costs
Variable Cost
t1 t2 Output / Time
Effects on Profit of Reducing Chance Only
PEW SOLUTION: Build a Life Cycle System that Creates
Plant and Equipment Reliability Improvement
Fewer profits
$ Accumulated Wasted Variable lost, but ‘fire-
Revenue
and Failure Costs fighting’ is
high
Total Cost
Fixed Cost
Wasted Fixed Costs
Variable Cost
Effects on Profitability of Reducing Consequence Only
Fewer
$ Profits
Revenue
Lost
Total Cost
Full Highly Lifetime
High Quality
Production Reliable Health and
Maintenance
Capacity Machinery Fitness Fixed Cost
Wasted Fixed Costs
Variable Cost
t1 t2 Output / Time
Effects on Profit of Reducing Chance Only
Day 1
Foundations
• Why Machines Fail
• Reliability
• Risk
• Cost of Failure
• Human Error
• Life Cycle
Reliability Properties for Series Systems
• Series Systems 1 1 1 n
Rsystem= R1 x R2 x R3 …..Rn
R = 0.95 x 0.95 = 0.9025
Reliability Properties for Series Systems
Rsystem= R1 x R2 x R3 …Rn
• Implications for Series Systems
1 1 1 n 1 System-wide improvements lift reliability higher
than local improvements. This is why SOP’s, training
and up-skilling pay-off.
Properties of Series Systems 2 Improve the least reliable parts of the least reliable
1. The reliability of a series system can be no
higher than the least reliable component. equipment first.
3 Carry spares for series systems and keep the
2. If ‘k’ more items are added into a series system reliability of the spares high.
of items (say 1 added to a system of 2, each
with R = 0.9) the reliability of all items must 4 Standardise components so fewer spares are
rise an equal proportion (3.5%), to maintain needed.
the original system reliability. 5 Removing failure modes lifts system reliability.
(0.9 x 0.9 = 0.932 x 0.932 x 0.932 = 0.81)
This is why Root Cause Failure Analysis (RCFA) and
3. A small rise in reliability of all items (say R of Failure Mode and Effects Analysis (FMEA) pay off.
the three items rises 0.93 to 0.95, 2.2% 6 Provide pseudo-parallel equipment by providing
improvement) causes a much larger rise in tie-in locations for emergency equipment .
system reliability (from 0.81 to 0.86, 6.5%).
7 Simplify, simplify, simplify – fewer components
means higher reliability.
Reliability Properties for Parallel Systems
• Parallel Systems 1
Rsystem= 1-[(1- R1)x(1- R2)x(1-R3)…x(1-Rn)] 1

(only for fully active)
R = 1 – [(1 - 0.6) x (1 - 0.6)] = 0.84
Reliability Properties for Parallel Systems
Rsystem= 1-[(1- R1)x(1- R2)x …(1-Rn)]
• Implications of Parallel
1 1 1 n
Systems for Equipment
1 Use parallel arrangements when the risk of
failure has high DAFT Cost consequences.
Properties of Parallel Systems 2 Consider providing various paths for product
1. The more number of components in parallel the to take in production plants with in-series
higher the system reliability. equipment.
2. The reliability of the parallel arrangement is 3 Build redundancy into your systems so there
higher than the reliability of the most reliable is more than one way to do a thing.
component.
m m m m
m m m m
Which arrangement is more reliable if m = 0.9?

What percentage improvement is the more reliable?
All Our Machines are Made of Parts in Series
Reliability = the chance of success
5
4 Shaft 6 Shaft 8 Lock 1 Inner 2 Roller 3 Outer
14 Lube 14 Lube Housing
Journal Seal Nut race bearing race Bore
R4 R6 R8 R1 R14 R2 R14 R3 R5
12 NOTE:
• Rn = Component reliability = chance of success the part will work right.
11 • To maximize Rn specify the conditions that deliver maximum ‘chance of success’
10
9
13
7 This is a reliability block diagram. Think of the string of
3
4
2
1
8
5 6
parts as the machine’s DNA. If a ‘gene’ is faulty in the
DNA then cancer starts and grows to kill the whole body.
1. For the motor to be highly reliable every

bearing must be even more highly reliable.
2. For the bearing to be highly reliable each of
Electric motor drive its parts must be even more reliable.
end bearing 3. For every part to be reliable its design and
operating health must be risk-free.
All Our Work is Made of Activities in Series
The Job
Rjob=
Task 1 Task 2 Task 3 Task 4 Task 5

R1 x R 2 x R3 x R 4 x R5
Rjob= Rtask1 x Rtask2 x Rtask3 x Rtask4 x Rtask5
NOTE:
• Rtaskn = task reliability = chance of success the task will be done right
• To maximize Rtaskn specify the conditions that deliver maximum ‘chance of success’

All Our Businesses are Processes in Series
The Business
Rbusiness=
Receivables Manufacture Assembly Despatch Customer
Rbusiness = Rprocess1 x Rprocess2 x Rprocess3 x … x Rprocess‘n’
NOTE:
• Rprocessn = process reliability = chance of success the process will be done right
• To maximize Rprocessn specify the conditions that deliver maximum ‘chance of success’

The Business and Supply Chains that Bind Us
Hundreds of activities across dozens of processes – what is your ‘chance of success’?
s2 s 10
Proces Proces
Proces
Proces
ProcesProcess 13 Proces
Process 19
Proces
Proces
Proces
Proces
s 25
s 13 Process 16 s 14s 6
s 16
s 17
s1
s6
s9
Process 2 Process 10 Proces s 23Process 23
Proces
Proces Proces
Process 25
Process 17 Process 19
s 24
Process 1
Process 6
Process 9
s5 Processs 1224
Ra Pre Ma Ass Pa Pr
Process 5 wR 12 parR
Process nufR
s2 em R cka R
s 10 od
Proces
Proces
Proces
Proces
Proces
Proces
Ma ati act Process 11
bly gin uct
Process 19
Proces
s 12
s 15s 9
s3
s7
s 8s 1
Proces
Proces
Proces
Proces
Proces
1
ter Proces
on
2
ureProces
3 Proces
4
g Proces
5 s 22 Proces
s 25
s 13
s 16
s 17
Proces
Proces
Proces
ials s 4 s 26 s 18
Proces s 23
s 20
s 14
Proces
Proces Proces s 24
Raw Preparation Manufacture Assembly
Proces s 5 Packaging s 12 Product
Materials s 21 Ra Pre Ma Ass Pa Pr
wR par
R nuf
R emR ckaR od
Proces
Proces Process
Proces
Proces
Proces
Proces
s 11
R1 R2 R3 R4 Ma ati R5 act bly gin uct
s 14
s 12
s 15
s3
s 7 Proces
s8
1
ter Proces
on
2
ureProces
3 Proces
s2
4
g Proces
5 s 22
s 10
Proces
s4 s 26 s 18
Proces
Process 14
ials
s 20
s 14
Process 12
Process 3
Process 7
Process 8
Proces Proces
Proces
Proces
Proces
Process 19
Proces
Proces
Proces
Process 4 Process 26 Proces Process 18 Proces
s 25
s 13
s 16
s 17
s 1 15
s6
s9
Proces s 23
s 21 Proces
Proces
Process 11 Proces s 24
s5 s 12
Ra Pre Ma Ass Pa Pr
Process 20
Process
wR parR nuf
R em R cka
R od
Proces
Proces
Proces
Process 22
Proces
Proces
Proces
s 11
Ma ati act bly gin uct
s 14
s 12
s 15
s3
s 7 Proces
s8
1 Proces 2
ureProces
3 Proces 4 s 22
g Proces
5
14
Process 21 ter on
Proces
ials s4 s 26 Proces s 18
s 20
s 14
Process 21
NOTE:
• To maximize the ‘chance of success’ each process will be done right, specify those conditions
that deliver maximum ‘chance of success’
Series System Reliability Property #1
1. The reliability of a series system can be no higher than the least reliable component.
R1 R2 Rsystem
0.9 0.8 0.72 Reliability with ‘Bad Actor’
12.5% 12.5% Percentage Improved
0.9 0.9 0.81 Reliability after Improvement
Local reliability improvements improve the system by the same amount.

RCA
RCA 0.9 0.8 0.9 0.9 0.58 Reliability with ‘Bad Actor’
RCA 12.5% 12.5% Percentage Improved
0.9 0.9 0.9 0.9 0.656 Reliability after Improvement
Series System Reliability Property #2
2. If ‘k’ more items are added into a series system of items the reliability of all items
must rise an equal proportion to maintain the original system reliability.
R1 R2 R3 Rsystem
0.9 0.9 0.81 Before adding one more item
-10%
0.9 0.9 0.9 0.73 After adding one item
3.45% 3.45% 3.45% +10%
0.932 0.932 0.932 0.81 After improving the system
Property 2 advises ‘Simplify, Simplify, Simplify’
13
11 12 14 5
10
Shaft 9
1 1
3 4 2
5 6
2 7
8 3
4
14 things to go wrong… 5 things to go wrong…
PEW SOLUTION: Apply Series System Reliability
Property #3 (best practise throughout the operation)
3. A small rise in reliability of all items causes a much larger rise in system reliability
R1 R2 R3 Rsystem
0.93 0.93 0.93 0.805
2.1% 2.1% 2.1% 6.5%
0.95 0.95 0.95 0.857
System-wide reliability improvements improve system reliability by a vastly greater amount.
0.93 0.93 0.93 0.93 0.748 Reliability before BEST PRACTICES

2.1% 2.1% 2.1% 2.1% 8.9% Percentage Improved BEST PRACTICES
0.95 0.95 0.95 0.95 0.815 Reliability after BEST PRACTICES
Day 1
Foundations
• Reliability
• Risk
• Cost of Failure
• Human Error
• Life Cycle
Common Wrongs Humans Do To Machines
• We twist them
• We squeeze them
• We hit them
• We poison them
• We burn them
• We shake them
• We break them
• We choke them
• We overload them
• We boil them
The Odds are Against Doing it Right!
Only one way to

disassemble
40,000+ ways to
incorrectly reassemble!
(Factorial 9 possible errors)
Read the article “The Human Factor” in the workbook
Source: Federal Aviation Authority Human Error Presentation, 2008
80% of Failure Events are Caused by
Human Factors and Human Error
Nuclear Plants: 88 –72%
Barringer, H. Paul, ‘Use Crow-AMSAA Reliability
Growth Plots To Forecast Future System Failures’
Boiler Accidents: 82%

(USA, ASME Report 2002)
Dams: 90%
(Henry, P., ‘Design Paradigms: Case Histories of
Error and Judgment in Engineering’)
Human Factors – limitations of Flesh and Bone
Physical Psychological
• Size • Experience
• Gender • Knowledge
• Age • Training
• Strength • Attitude
• The five senses • Emotional state
Physiological Psychosocial
• Health • Interpersonal relations
• Nutrition • Ability to communicate
• Lifestyle • Empathy
• Alertness/fatigue • Leadership
• Chemical dependency
Source: ‘Maintenance Human Factors Presentation,’ Federal Aviation Administration, 2008
The 12 Most Common Causes of Human Error
• Eliminate these causes and

you have conquered most
human errors
• Talk about each of these
• What are the causes in your

company?
• What are the corrective

actions?
PEW SOLUTION: Apply the Answers in the Human
Error Rate Table to Reduce Human Error 10,000%
~5 sigma
~4 sigma
1
3
2
~4.5 sigma
1
2
~2 - 3 sigma
3
Source: Smith, David J., ‘Reliability, Maintainability and Risk’,

Appendix 6, Seventh Edition, Elsevier – Butterworth Heinemann
The Table confirms that ‘human element’ error is real and unavoidable. We do not perform well
when tasks are structured in ways that require care and we perform especially badly under
complicated non-routine conditions. Add stress into that that mix and you get disaster.
Machines Suffer High Risk from Human Contact
Job 1 Job 2 Job 3 Job 4 Job 5 Job 6 Job 7 Job 8
Rjob= Rjob= Rjob= Rjob= Rjob= Rjob= Rjob= Rjob=
R1 x R2 x R3x R4 R1 x R2 x R3x R4 R1 x R2 x R3x R4 R1 x R2 x R3x R4 R1 x R2 x R3x R4 R1 x R2 x R3x R4 R1 x R2 x R3x R4 R1 x R2 x R3x R4

x R5 x R5 x R5 x R5 x R5 x R5 x R5 x R5
Task 1 Task 2 Task 1 Task 3 Task 2 Task 41 Task 3 Task
Task52 Task
Task41 Task 3 Task
Task5 2 Task
Task41 Task 3 Task
Task
5 2 Task
Task41 Task 3 Task
Task5 2 Task
Task41 Task 3 Task
Task5 2 Task
Task14 Task 3 Task
Task 25 Task 4 Task 3 Task 5 Task 4 Task 5
Shaft Shaft Lock Inner Roller in Outer Housing

Parts shown
Lube Lube
Journal Seal Nut race bearing race Bore as a series
Shaft
Journal
Shaft
Seal
Lock
Nut
Inner
Race
X Roller
bearing
X Outer
race
Housing
Bore
“One fails; all fails”
“One poor; all poor”
Reliability of Series Work Process
Series Tasks
1 2 3 4 5 6 7 8 9 10 11 12
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12
1 1 1 1 1 1 .9 1 1 .9 1 1
.9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9
.99 .99 .99 .99 .99 .99 .99 .99 .99 .99 .99 .99
Top System Mid System Bottom System
R system= R1 x R2 x R3 … R system= R1 x R2 x R3 …
R system= R1 x R2 x R3 …
0.9 x 0.9 x … 0.9 0.99 x 0.99 x … 0.99
0.9 x 0.9 x 110
(0.9)12 = 0.2824 (0.99)12 = 0.8864
0.81
If 50 tasks (0.9)50 = 0.0052 If 50 tasks (0.99)50 = 0.6
Carpenter’s creed: ‘measure twice, cut once’
Get wood Measure 1 Mark wood Cut wood 1 error every 200
opportunities
R= 0.995 ~ 1 / wk
0.995
Get wood Mark wood Cut wood 1 error every 5000
0.995 opportunities
~1 / 20 wk
??? ??? ???
PEW SOLUTION: Use the power of parallel
proof-tests on every task activtity
Original task
reliability 0.9 0.9 0.9 0.9 0.9 0.59
AND
Proof-test 0.99 0.99 0.99 0.99 0.99 0.95
reliability
R system= 1- [(1- R1)x(1- R2)x(1-R3) …]
1- [(1- 0.9)x(1- 0.99)]
1- [0.1 x 0.01]
1- [0.001] = 0.999
Equivalent
series 0.999 0.999 0.999 0.999 0.999 0.995
reliability
Can we get 10,000% fewer errors?
(Two sigma to five sigma)
Every Maintenance Job … …is a Series Work Process
Work Environment
1 2 3 4 5 6 7 8 9 10 11 12
R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12
Each task can be made more certain if we can include

redundancy and turn it into a parallel arrangement.
Work Environment
1 2 3 4 5 6 7 8 9 10 11 12
1 2 3 4 5 6 7 8 9 10 11 12
Parallel Proof Test Activity Reliability
Parallel Tasks
.9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9
.9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9
Equivalent Series Tasks

.99 .99 .99 .99 .99 .99 .99 .99 .99 .99 .99 .99
Parallel Proof Test Reliability Series Tasks with Parallel Test Without Parallel Test
R system= 1- [(1- R1)x(1- R2)x(1-R3) ….] R system= R1 x R2 x R3 …. R system= R1 x R2 x R3 …

1- [(1- 0.9)x(1- 0.9)] 0.99 x 0.99 x … 0.99 0.9 x 0.9 x … 0.9
1- [0.1 x 0.1] (0.99)12 = 0.8864 (0.9)12 = 0.2824
1- [0.01] = 0.99
If 50 tasks (0.99)50 = 0.6 If 50 tasks (0.9)50 = 0.0052
Clearly, adding a proof test into each task to check accuracy makes tremendous improvement in
workmanship quality. This process went from a 28% chance of it being done right, to a 89% chance!
How Much Must We Control Chance?
Add in more Parallel Check-Test Tasks
.9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9
.9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9
.9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9 .9
Equivalent Series Tasks

.999 .999 .999 .999 .999 .999 .999 .999 .999 .999 .999 .999
Task With Second Parallel Proof Test Series with Second Parallel Prrof Test How many 9’s is enough?
R system= 1- [(1- R1)x(1- R2)x(1-R3) ….] R system= R1 x R2 x R3 …. R system= R1 x R2 x R3 …

1- [(1- 0.9)x(1- 0.9) x(1- 0.9)] 0.999 x 0.999 x … 0.999 0.9999 x … 0.9999
1- [0.1 x 0.1 x 0.1] (0.999)12 = 0.988 (0.9999)12 = 0.999
1- [0.001] = 0.999 1 failure in 1000 opportunities
If 50 tasks (0.999)50 = 0.95
If 50 tasks (0.9999)50 = 0.995
5 failures in 1000 opportunities
Adding tests and checks reduces the chance of failure.
BUT… beware of common cause error across tests.
But Parallel Test Tasks adds Cost…
So what can you do?
BEST ANSWER IS TO USE ‘LEAN ERROR PROOFING’,… IF IT IS WRONG
THE ERROR IS FOUND AND CORRECTED SO IT DOES NO HARM.
1 1 1 1 1 1 1 1 1 1 1 1
ERROR PROOFED Series
R system= R1 x R2 x R3 ….
1x1x…1
(1)12 = 1
If 50 tasks (1)50 = 1
A PERFECT OUTCOME EVERY TIME!

Day 1
Foundations
• Reliability
• Risk
• Cost of Failure
• Human Error
• Life Cycle
What Makes a Productive Equipment Life?
MAINTENANCE KPI: Unit Cost
When you make plant
more reliable you
Maintenance proportion = Cost
High Capacity
work on the ‘capacity’
of the Unit Cost
Return part of the Unit Cost
On Investment equation. As a result
you drive down the
High Productivity, cost of your product
Low Operating Cost because the plant is
available to work at
High Availability, full capacity for longer.
You make more
High Capacity product in the same
time for less cost.
High Reliability
Robust, Built & Operated Maintain to Continually

Suitable Installed Within Design Improved
Design Correctly Limits Standard
The Life Cycle of Plant and Equipment
Equipment Life Cycle
Project Phase of Life Cycle Productive Phase of Life Cycle End
Decommissioning
Disposal
Feasibility
Approval
Commissioning
Preliminary Design
Idea Creation
Detail Design
Procurement
Operation
Construction
1 1 1 n Profits come from this stage of the life

cycle, and are maximised when the
Life Cycle is a Series Process operating costs are minimised.
Rlife cycle = R1 x R2 x R3 … Rn
Effect of Process Failures Across Life Cycle
Process failures during this phase will cause plant and equipment failures in operation.
Equipment Life Cycle (say 20 years)
~ 10% of Life Cycle (~ 2 years) ~ 85% of Life Cycle (~ 17 years) ~ 5%
Decommissioning
Feasibility
Disposal
Approval
Commissioning
Detail Design
Procurement
Preliminary Design
Operation
Idea Creation
Construction
System
Chance
of Failing
Component
Chance of
Failing
When Operating Costs are Committed
Once a plant is
designed and built
there is very little that
can be done to reduce
operating costs
because they are
substantially fixed by
the plant’s design. If
you want low
operating costs, this
chart makes it clear
that they are designed
into the plant and
equipment during
feasibility, design and
construction.
Low Operating Costs are Designed-in for Life
Organisations separate projects from operations and unwittingly create a divide between groups.
You must create continuity of information back and forth along the life cycle so the total life cycle
costs are minimised. Low capital costs do not necessarily mean most profit in future.
Equipment Life Cycle (say 20 years)
~ 10% of Life Cycle (~ 2 years) ~ 85% of Life Cycle (~ 17 years) ~ 5%
Decommissioning
Disposal
Feasibility
Commissioning
Approval
Detail Design
Procurement
Preliminary Design
Operation
Idea Creation
Construction
Specify the
Production Needs …
Capital Project“To
~ 10be- 20%
… of Life Cycle Cost Operation ~ 80 - 90% of Life Cycle Cost
On Time
A proportionately small capital expenditure in the project phase on better equipment and strategies
On Quality
On Cost
(like intentionally buying bigger equipment and de-rating it to lessen the internal stresses) lowers
On Profit
operating costs, and …consequently delivers higher operating profits all the years of the plant’s life.
“… we need 98% 87
Plant-wide Availability!”
PEW SOLUTION: Stop Variability and Defects Across
the Business and Plant and Equipment Life Cycles
Every process throughout the life cycle will create many defects.
Management Engineering Supply Contractor Operations Maintenance
Install Operate
Maintain
Specify Design Store
Buy Start-up
Introduced
Defect and
Failure Cost
Surge
Operating Plant Uptime defects
and Throughput
Variability in
each process
causes defects
1 Serious Failure
which at times
10 Losses Product progress to
6,500 Repairs
Higher Unit Cost, ‘failure’.
20,000 Defect Modes
Poor Quality and
The Failure Pyramid
Source: Ledet, Winston, The Manufacturing Game
Delayed Delivery Source: Thanks to Ron Moore from Ron Moore
Group in the USA for this concept.
Manage and Continually Reduce the Risks to
Your Plant and Equipment Life Cycle
Equipment Life Cycle
Project Phase of Life Cycle Productive Phase of Life Cycle End

Preliminary Design
Decommissioning
Feasibility
Commissioning
Approval
Detail Design
Procurement
Operation
Idea Creation
Construction
Disposal
To be sure we consider the long-term well-being of our business, we view the
health of its plant and equipment from the end of their life cycle. We ask what do
we do to get a long, healthy, problem-free operating life?
So by selecting methods and practices that lead to lower

risks and trouble-free operation, we also maximise life
cycle profit because there are far fewer DAFT Costs to pay.
PEW SOLUTION: Life Cycle Risk Management Strategy
(DOCTOR) Optimised Operating Profit Method
Profit Optimisation Loop

It is possible
to make great
operating Assume DAFT
Design Redesign for
cost savings Equipment Costs
Drawings Lower Risk;
during Failure Spreadsheet
Revise
design, if the
Projected O&M
designers Failure Cost N
R&M Strategies;
reduce the Acceptable?
Costs Revise
operating
Project
risks that Y
Busine$$ Strategies
their choices
cause the Ri$k Ba$ed Frequency N
business. Equipment Achievable?
Criticality
Y Quality Procedures
Precision Maint
FMEA/RCM
HAZOP Applicable Applicable Predictive Maint
Preventive Maint
Precision Standards Project O&M RCFA
Precision Instaln
Reliability Eng Strategies Strategies Maint Planning
Etc.
Etc.
Day 1
Foundations
• Reliability
• Risk
• Human Error
• Life Cycle
How do we to Stop Problems Happening?
Maintenance sees … Production sees …
What We See
What Caused It
To getTothe
getright
the end
rightresults
end results
we first
youneed
firstto
need
get to
theget
basics
the
right.
foundation
Then the basics
right
right.
results
Thencanthe
follow.
right outcomes can follow.
Journey to 6 Sigma: Minimize Variability
Accuracy Controlled
Enterprise (ACE 3T)
Total Quality Control (SPC)
Trained People with Management
‘Living’ ISO 9001 QMS

Performance Levels
Perfect processes
Systematize processes;
Standardize procedures
Fix the obvious;

Do basics well
1σ 2σ 3σ 4σ 5σ ‘6 Sigma’ Rating
69 31 7 0.6 0.023 Errors per 100
Sources: David Burns, SIRF Roundtables Ltd, Melbourne, Australia
George. Mike et al, ‘What is Lean Six Sigma’, McGraw-Hill, 2004
Chance of Trouble at Each Sigma Level
Probability of Process Six Sigma
Ranking Failure Rate Error Rate
Failure Capability Rating
Unlikely 1 4 in 10,000,000 ± 5σ 7 2 in 100,000,000
Very Low 2 1 in 16,000 ± 4σ 6 3.4 in 1,000,000
Low 3 1 in 2,000 ± 3.5σ 5 2.3 in 10,000
Moderate 4 1 in 1,000 ± 3.3σ 4 6 in 1,000
5 1 in 500 ± 2.65σ* 3 6.7 in 100
6 1 in 100 ± 2.5σ* 2 31 in 100
High 7 1 in 50 1 69 in 100
8 1 in 25
Very High 9 1 in 10
10 1 in 5
* = process not capable at minimal levels
Source: Dovich, Robert., ‘Perspectives on Quality, FMEA and You’, Volume 1, No 7.
Most Businesses make their Machines Break
MAINTENANCE KPI
Breakdown Hours
Hours Control Chart
Too many Major Failures (Outliers)
± 3 sigma
Week No
This is a statistically stable process of breakdown creation –

this business makes breakdowns as one of its ‘products’.
PEW SOLUTION: “Chop out the Roots of Failure”
Get rid of the cause of poor Get rid of the cause of
REACTIVE MAINTENANCE
connections!
X
vibration!
X
X
Get rid of the cause of oil
contamination!
“Chop Out “Chop Out

the Roots” the Roots”
PROACTIVE MAINTENANCE
PEW SOLUTION: Use Standardised Work Procedures
to Limit Task Variability to Within Quality Standards
First draft of standard/procedure
First review and

improvement
Time
First trial
Range of Possible Outcomes for Time,
Quality Cost, Quality
Second draft
Second trial
Standardised Procedure
means… developing one best way Third draft
Third trial
±10% Range of Possible Outcomes for
Time, Cost, Quality
New
New Ideas Procedure Technology
Remove the Variability in a Job by Using 3T
(Target-Tolerance-Test) Error Proof Techniques
Only accept this range of outcomes
because they give very low risk
Number of Events
Very Bad Outcome Very Bad

Outcome
Acceptable
Outcome
Range of Values of a Critical Parameter
By setting quality controls into a job you ensure the actions that create reliability are
done thereby greatly reducing the chance that a mistake will be made.
In the end... reliability is a quality control issue because

the standards you meet create the reliability you get
Why Setting Standards is Vital:
Lubricating Oil and Hydraulic Oil Cleanliness
Source: Wayne Bissett, ‘Management of Particulate Contamination in Lubrication Systems’ Presentation,

CM and Lubrication Forum 2008, Australia
The Importance of Setting Standards
Case Study 1 : Shaft Quality Control for Bearing
Reliability
See the example in the workbook on how to check the
quality of a shaft is suitable for a roller bearing
From the manufacturer or appropriate

engineering specifications
In Each Work Task Provide a Target to Hit
Goal Posts and Targets

•Set a target of what is the best performance and people will aim for it!
• Goal posts and targets set requirements for excellence.
• SOPs need targets to meet as part of each task instruction.
Provide a Tolerance Range of Performance
29.8 to 30.2 is
acceptable
On an archery target you

know exactly how well you
have done! You should also
know how well you have
done your job.
Considered accurate if within tolerance – Specify a range of acceptance.

Provide a Test For Proof of Accuracy
Appling MEASUREMENT to TEST and PROVE accuracy

produces self-correcting learning. www.lifetime-reliability.com
PEW SOLUTION: Develop and use Accuracy
Controlled Enterprise (ACE) 3T Procedures
“An organisation that applies error proof 3T accuracy
controlled procedures at every level of operation.”
Task Task Task Record Action if

Sign-off After
Step Step Step Full Description of Task Test for Correctness Actual Out of Complete
No. Owner Name Result
Tolerance
(Max 3 – (Include all tables, diagrams and pictures here) Describe the test and below specify the
4 words) target as ‘BEST’ and range of acceptable
results that are ‘Good enough’.
Good Better Best

1
Tell people
how to fix the
problem
One Layout for an Accuracy Controlled 3T – Target,

Tolerance, Test – Failure Prevention Procedure
PEW SOLUTION: Using 3T’s to improve Supplier
Performance and Craft Workmanship Quality
As MAGNIFICENT as it needs to be As BAD as you are allowed
BEST BETTER GOOD
PASS / FAIL /
ACCEPT REJECT
Perfect World Class Tolerance Certain

Result Target Limit Failure
What is Good, Better, Best

for setting your clock?
PEW SOLUTION: Develop and Use Reliability
Creating ACE 3T Error Proof Procedures
Build ACE 3T Mistake Proofing into SOPs
•Set a target for each task.
•Specify the acceptable tolerance.
•Do a test to prove accuracy.
Bands of Lesser
Frequency of Outcome
Quality
Quality
(Decreasing Accuracy) improvement
occurs when
‘Good, Better, Best’ variation is
‘Bronze, Silver, Gold’ reduced
Output
Specification
Range of Possible Outcomes
Use Two-Sided Error Proof Quality Control
• Put Mistake Proof Controls into SOPs with clear 3T task standards
•Use the As-Designed Standards… then improve them
• Work to Internationally Recognised Standards Accuracy
– Set a Target for each task Precision
– Specify acceptable Tolerance range Best

Better
– Do a proof Test to prove accuracy
Measure
Good
World-Class BEST
Standard
BETTER
Stretch Standard
Specification
Equipment GOOD Range of Outcomes
Manufacturer’s “By stating the exact requirements to meet you are causing
Standard people to take all necessary actions to achieve them.”
Write SOPs that include Failure Prevention
Sign-off
Task Task Task Record Action if
Tools & Full Description Test for After
Step Step Step
Condition of Task
Tolerance Range Actual Out of Complet
No. Owner Name Correctness Result Tolerance e
(Max 3 (Include all tables, Good Better Best

– 4 diagrams and
words) pictures here)
Pass / World Tell people

Fail Class how to fix the
OEM Half What the problem
spec is way very best
least stretch achieve
allowed goal
When procedures are written with the 3Ts you can guide people right to the outcome they need
to deliver. We build into 3T procedures the necessary actions, that when performed, will deliver
the maintenance strategy. We give people a way to check that their work is exactly what it
needs to be. They self-improve and gain the self-satisfaction of having done a great job.
Continuous Improvement with Accuracy
Controlled Procedures
Now you know what ‘good enough’ is!
Task Task Task Record Action if Sign-off

Tools & Full Description of Test for
Step Step Step Tolerance Range Actual Out of After
Condition Task Correctness
No. Owner Name Result Tolerance Complete
(Max 3 – (Include all tables, Good Better Best

4 words) diagrams and pictures
here)
Drives Improvement
Accuracy Controlled 3T Procedure Layout
Continuous improvement: Make ‘better’ ‘good’, make ‘best’ ‘better’ and set a new
standard for ‘best’. In this way, you will drive quality improvement and innovation in
your company. ‘Good enough never is!’
Activity – Put 3T Quality Control in Procedures
NOTE:
None of these task will
prevent the pump and
piping from failing. These
tasks find future repairs…
and you want a healthy,
reliable pump set...
…prevention is easier than
cure
Precision Maintenance of Machinery is …
1. Accurate Fits and Tolerance at Operating Temperature
2. Impeccably Clean, Contaminant-Free Lubricant Life-long
3. Distortion-Free Equipment for its Entire Lifetime
4. Shafts, Couplings and Bearings Running True to Centre
5. Forces and Loads into Rigid Mounts and Supports
6. Laser Accurate Alignment of Shafts at Operating Temperature
7. High Quality Balancing of Rotating Parts
8. Low Total Machine Vibration
9. Correct Torques and Tensions in all Components
10. Correct Tools in the Condition to do the Task Precisely
Number 14 is the one that the vast
majority of companies miss. They
11. Only In-specification Parts don’t systemize and standardize
the delivery of precision to their
12. Failure Cause Removal during Maintenance machinery.
13. Proof Test for Precision Work and Equipment

14. A quality system to make all the above happen 112
PEW SOLUTION: Outstandingly Reliable
Machines Require…
REQUIREMENT TARGET TOLERANCE
1. Chemically correct, Contaminant-Free Lubricant Right Viscosity; <100ppm water; ISO 12/9 <ISO 14/11
2. Accurate Fits and Tolerance at Operating Temperature Form IT5, Temperature to design <IT7
3. Shafts, Bearings and Couplings Running True to Centre IT5 <IT7
4. Distortion-Free Equipment for its Entire Lifetime IT5 <IT7
5. Forces and Loads into Rigid Mounts and Supports No Looseness; Safely absorb/dampen forces
6. Accurate Alignment of Shafts at Operating Temperature Coupling/Feet offset 10µm/20µm <20µm/40µm
7. High Quality Balanced Rotating Parts G1 <G2.5
8. Total Machine Vibration Low 1.5mm/s rms <3mm/s
9. Correct Torques and Tensions in all Components ±5% of correct tension <±10%
10. Correct Tools in Precise Condition to do Task to Standard As new condition/calibrated
11. Only In-specification Parts OEM approved material and design specs
12. Failure Cause Removal during Maintenance Creative Disassembly and Precision Assembly
13. Proof Test for Precision Assembly Quality Every task proven correct Milestone Tests
14. A quality assurance system to make all above happen ACE 4T ITP
(Inspection & Test Plan)
NOTE: These parameters are indicative and may not apply to a particular machine. Confirm actual requirements with the manufacturer.
PEW SOLUTION: World Class Standards produce
World Class Equipment Reliability
Source: Wayne Bissett, OneSteel
Reliability Manager, Planning and
Condition Management Presentation,
Sydney, Australia, 2008
‘Precise’
‘Smooth’
‘Tight’
‘Dry’
‘Clean’
‘Cool’
‘Repeatable’
“Only world class standards can produce world class results.”
Asset Management in a Nutshell
Business RISK performance of an asset is influenced by two things:
– Its inherent reliability established at the design stage.
– Its achieved availability, which depends on how well the asset is
operated and maintained through its life.
MAINTENANCE:
•Contributes to the success of the organisation
•It’s purpose is to eliminate the need for maintenance
•It’s goal is to do everything in its power to
– Increase the quantity and quality of production, and
– Reduce costs BY REDUCING THE CHANCE OF RISK
Sources: Jim Wardhaugh, UK Consultant (30 years with Shell
and Centre of Excellence Leader)
PEW SOLUTION: A View of the Plant Wellness
Way (PWW) Climb Up to Reliability Excellence
100%
Maximum
Availability Plant and
Equipment
Wellness
Business
Equipment Availability
Predictive Proactive Systems Focus

Maintenance Maintenance
Condition Monitor Equipment
for Start of Failure Reliability Focus
Preventive
Maintenance Business System
Upkeep
Limiting
Availability and
Maintenance Cost
Breakdown
Maintenance
After Failure
< 80%
Destructively Exceptionally
High Low
Maintenance Cost
Integrated System to Reliability & Plant Wellness
Unit Cost of Maintenance
Performance
MTBF WO per
Measures
Operating Profit Uptime

Production Compliance Technician
Frequency of Operational Technician per
ROI Zero Incidents Maintenance Cost
Overstress Events Overloads ERV
Operating Costs
Organizational
Asset Strategy
Assets Assets’ Assets’ Assets’ Assets’ Assets’

Financial Operational Performance Reliability Engineering Maintenance
Objectives Strategy Objectives Objectives Standards Objectives
Marketing Assets’ Assets’

Objectives Operating Maintenance
Standards Strategies
Periodic Audit
Action Plan
and Training
Plan
Continuous Financial and Implementation

Improvement Performance Plan
Measures
PEW Enterprise Asset Management Model

Need System Wide + Life Cycle Wide Solutions
We want This needs a

maximum plant management
and equipment system
life cycle profits over the whole
life cycle
A Asset Management System that coordinates the Organisation

Design Buy Store Install Operate Maintain
CEO
Production Engineering Marketing Admin
Shopfloor Maintenance
Supply
World Class Practices Required at Every Step
Use A Quality Management System incorporating Proof Tests to drive
Accuracy and Continuous Improvement
Specify Design Buy Build Run Maintain
Risk Risk Quality Precision Precision Precision

Removal Removal Control

3T
Plan
 
Do
3Ts are used here to  
3T guarantee improved performance Check
(Measure)
reliability in every process
 
Act
(Improve) 
3T
Follow the Processes of Plant Wellness Way
We can
summarise the
key
requirements in
a simple chart
Mindset to show people
Barrier the necessary
philosophies
that need to be
introduced at
Skills/Knowledge the different
Barrier parts of ‘The
Journey’.
Companies
often hit
‘barriers’ and
they struggle to
get past due to
old paradigms.
Develop the People of Plant Wellness
People want to
feel valuable, and
to be valued. You
can do both
when you
increase their
worth to a
business. This
diagram shows
where you need
to take your
people, and as a
consequence
they will bring
your business
with them.
Overcoming the ‘Skill and Knowledge Barrier’
Observe the Financial Steps to Plant Wellness
Don’t forget that
your financial
recording and
analysis systems
will need to keep
pace with the
informational
requirements the
Plant Wellness
philosophies.
Teach the Leadership Culture of Plant Wellness
Culture is ‘the
way we do things
around here’.
Overcoming the ‘Mindset Barrier’
You want a
culture of
continual
improvement,
where people
naturally come
up with new
ideas and are
helped to trial
them. You don’t
want the culture
of ‘that’s the way
we’ve always
done things
around here’.
PEW SOLUTION: The ‘Wholeness’ of Plant Wellness
People, money,
machines and
processes all
working together in
an integrated
Processes People learning system,
where business life
cycle risk is
controlled to
maximize machine
and equipment
health and
Capital Culture wellbeing, so the
business and its
people profit – that
is ‘Plant Wellness’.
PEW SOLUTION: The Plant and Equipment
Wellness Way to Operational Excellence
Your Ideal Operational
. Excellence Your Ideal Operational Excellence
Asset Management System Asset Management System
Engineering, Maintenance, Precision Quality Operation
and Operational Life Cycle Standards Management Standards
Quality Engineering
Quality Processes Standards Standards
Construction, Precision Engineering
Operations and Skills Materials Design Operator
of
Maintenance Practices Construction Skills
Defect Environment
Elimination Stress
Precision ReductionPrecision
Strategy Installation Operation
Operating Distortion Degradation
Risk Control Control
Reduction ATOMIC STRESS TO
Machine BUSINESS PROCESS MODEL Component
START HERE Parts Distress
for world class results Health (Atomic Structure
www.lifetime-reliability.com Failure)
PEW SOLUTION: Do the Six IONICS Processes to
Plant and Equipment Wellness
6-STEP PLANT WELLNESS PLANT WELLNESS
METHODOLOGY IONICS PROCESSES
Measure the total cost of your 1 – Identify Risks:

operating risks. Operating Risk
(How much will you lose if things go wrong?)
Identification Process
Which risks will you accept and 2 – Order Risk by Importance:

which will you prevent? Operating Risk Selection
(What risk boundary will you accept?)
Process
What do you do to prevent 3 – Numerate Options:

unacceptable risks and Risk Control Planning
encourage good results?
Continual Improvement Loop
Process
How do you make changes that 4 – Introduce Risk Reductions:

prevent risks and increase good Risk Control Introduction Process
outcomes?
Are the risk reduction changes 5 – Controlled Processes:

working? Monitoring and Measuring
Process
How do you improve what is not 6 – Synthesize New Ideas:

yet working well? Operating Risk Continual
Improvement Process
A LIFETIME OF HIGHLY
RELIABLE PRODUCTION PLANT
AND EQUIPMENT 126
The Plant Wellness Way: A Business Wide and
Life Cycle Long System of Asset, Operations
and Maintenance Management
Reliability Human Document & Learning Capital Project Business Risk
Engineering Capability Configuration Organisation Management Management
Management Management Business
Planning and Integrated Knowledge Process Life Cycle
Scheduling Business Systems Management Improvement Costing
ACE Precision ACE Quality ACE Quality ACE Engineering ACE Operation
Standards Standards Management Standards Standards
Supply Chain Lean Workplace Stores & Materials Operating Risk
Management Practices Management Reduction
Precision Robust Materials of Precision
Maintenance Engineering Construction Operator
Skills Design Selection Total Skills
Preventive Predictive Productive
Maintenance Maintenance Maintenance
ACE 3T Precision Proactive ACE 3T Precision
Installation Maintenance Operation
Parts’ Environment Operating
Management Stress Reduction
Distortion Degradation
Management Management
Failure Cause
Prevention
Equipment
Parts’ Health
START HERE (Low Atomic
for world class system Stress)
The PEW/PWW Strategy….
• Maximises life cycle profit, by
• Markedly reducing human error, and
• Proactively minimizing component stress, thru
• Greatly increasing work quality results, to
• Produce the world class reliability, that
• Delivers the least maintenance commitment, and
• Brings Operational Excellence performance, by
• Creating a single holistic business system, to
• Address life cycle cost issues and reduce variation, that
• Produces the least operating equipment risk profile, and
• Results in the least operating costs forevermore.

CMRP Training

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

CMRP Training

Загружено:

Авторское право:

Доступные форматы

The Plant and Equipment Wellness Way

to Enterprise Asset Management Success and

3-day training course

Day 1 Day 2 Day 3

2nd bearing sleeve 2nd bearing bush

1st bearing sleeve 1st bearing bush

Point contact only

Shaft misalignment distorts and

a part can experience at the contact points as a

Parts with least strength

Parts fail whose strength

Parts ‘age’ as they are used. Loads stress the

10,000 cycles Limited life at Infinite cycles

Using a Factor of Safety means the Designer

No of Stress Cycles Before Failing

Failure below the infinite life

20 cycles at this 200,000 cycles at Infinite cycles at

What it should be..

Negligible P. Low Risk Temperature Rise

Production 30m 50m 28m

Many parts fail

degradation; there is Another Smooth Running

The Stress-Driven Failure Degradation Sequence

Equipment replaced here – Few Problems! Time/Load Cycles

Time or Usage Age of Parts

What can cause this glass to break?

• It can be dropped, for example -

3. slip out of a bag or carry box

3. slip out of a bag or carry box

Time – Age of System

The Maintenance Zones of Equipment Life

To improve the reliability of a series of parts (that’s a machine) we must improve

Time or Usage Age of System

Remove Causes Time or Usage Age of Parts

Source: Wayne Bissett, OneSteel Reliability Manager, Planning and

Feel – Operator judgement ± 35% 1

Torque Wrench ± 25% 1.5

Load Indicating Washer ± 10% 3.5

- 15% + 15% Torque

Feel – Operator judgement ± 35% 1

Load Indicating Washer ± 10% 3.5

- 15% + 15% Torque

Risk = Consequence $ x Likelihood /yr

Must Stop Deformation

Only Excellent Only the Correct Only Proper Fits Absolutely No

How often will a failure Repair Business 0% 50% 100%

$100K $1,000K Outside the Volume Never Accept Failure

Risk $/yr = Consequence $ x Frequency of Failure /yr

Too many small failures is just A

Risk = Consequence x Frequency

1 10 100 1,000 10,000 100,000 Log of Consequence $000

Log Frequency No/yr

In reality, extreme risk

Negligible / Acceptable Risk

High RISK MATRIX

Actual Engineering, Ops and

50-70% 10-30% REQUIRED

Stress Removal Business Wide ACE 3T Lifetime

Life Cycle Operating Risk

MAINTENANCE OPERATIONS ENGINEERING

Event does occur CM oil condition analysis

Once every Almost

Profit ($) = Total Costs ($) =