S-RCM Awareness Training P&T R7a 23jul2010

SHELL RELIABILITY CENTERED MAINTENANCE
COURSE, LEVEL 1 - AWARENESS
Introduction into the concepts

of Shell Reliability Centred
Maintenance
Frat Amarra, Nagapati Hegde and Harry van Teijlingen

Reliability Management EU NL - PTE/OEXE
Property of Royal Dutch Shell plc RESTRICTED July 2010 1

Rijswijk – The Netherlands
Background notes: .
This course describes the aspects of Shell Reliability Centred Maintenance to
train people involved in the S-RCM process to Awareness competency level.
It involves the first steps to make one familiar with the methodology and how S-
RCM is relevant to the Business. Awareness competency means one will be able to:
• Describe the main elements of the S-RCM work process,
• Describe the importance to the business and
• Recognize what part of the work process is relevant to you.
This course is intended to all involved in the S-RCM methodology, e.g. Process
Operators, Technologists, Preventive Planners, Maintenance craft and Discipline
engineers. This training will prepare you to participate in S-RCM activities and
enables you an efficient involvement.
The course can also be useful for RCM specialists like Process focal points,
Process owners and S-RCM facilitators, as a starting point to develop further.
This web based training takes about 3 hours to complete.
At the end of this training a small quiz should be completed that will provide you
with your S-RCM Awareness Certificate.
Next, you need a piece of paper and a pen, Success !

Rijswijk – The Netherlands S-RCM Awareness
Dilemma: how to optimize?
Please take a sheet of paper and a pen, ok?
Imagine you just started in your job on a plant:

— Your boss indicates there are 2000 preventive maintenance tasks (PM‘s)
executed per year and he challenges this scope of work.
— Still there are often high priority repairs in daily maintenance and
plant reliability is not up to standard.
— He has asked you to improve this.
How would you do this?

— How to justify the PM‘s done on cost effectiveness?
— How to reduce high priority work?
— How to improve plant reliability?
Please write these answers down how to improve this.

Background notes: .
The traditional approach of reliability problems is often to apply more
maintenance, based on the break-downs found over the years.
This leads to a growing scope of preventive maintenance tasks, often also applied
for failures that happen rarely or for failures with negligible consequences.
However the same failure of an equipment can have high consequences if it is on a

critical piece of equipment in the plant.
To improve this the preventive maintenance tasks (PM‟s) should not only be cost
effective. For the PM‟s to be effective one should know:
• What degradations can happen?
• How does it degrade? What parts involved?
• Does the degradation and failure matter?
• What preventive maintenance tasks can be applied to mitigate the risk?
• Are these tasks cost effective?
This analyses are done in Shell Reliability Centered Maintenance.

Exercise on cost effectiveness of preventive maintenance:
In your plant, one of the high priority maintenance request involve a pump.
You asked specialists advice for preventive maintenance tasks, this is what
happened:
If the pump fails the consequences will be a loss of $ 100k.

If we do nothing it will fail every two years
What is the average loss per year of this FM??
A maintenance plan is recommended and costs $ 40k per year.

The consequence is that the pump might fail every 5 years.
Now: Do you think is this a good preventive maintenance plan? Y or N
Why?
Do you think this package adds value? How much?

Background notes: .
This is a simple example to determine cost effective maintenance.
Please calculate a cost effective approach: To execute the tasks or not?
It is recommended to choose a time frame, e.g. 1 year

( but it can also be 2 years or 5 years).
Then calculate the cost of not doing the maintenance. Please write down:
If we do not this maintenance the average cost per year is: ………………………
And when we do executing the preventive maintenance tasks,

the average cost per year is: ………………………
What is your choice?

How much is the benefit?

Exercise Answer:
If the pump fails the consequences will be a loss of: $ 100k.

If we do nothing it will fail every two years
The average loss per year in the ‗no maintenance situation‘ would be:
$ 100 k / 2 years = $ 50k/year
The maintenance plan costs $ 40k per year.

But: the consequence is that the pump till might fail every 5 years.
This equals to $ 100k / 5 years = $ 20k/year
Together the cost in the ‗maintenance‗ situation would be:

$ 40k/year + $ 20k/year = $ 60k/year
So:
without maintenance the average costs are lower than with maintenance.

Background notes: .
In this example the risk of failure is reduced from 50 k$/year to 20 k$/year, so
it looks like an effective set of preventive maintenance tasks.
However the costs of the maintenance tasks is relatively high,

also there is an significant residual risk.
This causes that it is more cost effective to do nothing then to do execute the
preventive maintenance plan.
The verification of the cost effectiveness is part of Shell Reliability Centered

Maintenance (S-RCM).
S-RCM helps to select cost effective preventive maintenance tasks and to

improve them.
In this case it could lead to two next steps:

1. Find less expensive preventive maintenance tasks,
2. Find preventive maintenance tasks that further reduce the probability of
failure.

Shell Reliability Centered Maintenance

Background notes: .
With process plants and equipment in operation, over time degradation occurs and
equipment failures can take place.
Some equipment failures can have small effects and others can have large effects
towards plant availability, health and safety and environmental aspects.
S-RCM analyses equipment failures, looks at the consequences of loosing

functionality and provides preventive maintenance tasks to deal with it in a cost
effective way.
S-RCM is one of the elements of reliability management of a process plant.

What is RCM about
RCM is about maintenance activities.

Maintenance is:
the combination of all technical and administrative actions
intended to retain an item in - or restore it to - a state
in which it can perform its intended functions.
 So: Maintenance is not a Department and maintenance actions
may involve both Maintenance and Operations employees.
RCM is a process to specify technically appropriate, technically

feasible and economically justified actions to maintain the functions of
equipment.
 These actions should preserve the functions, prevent premature
failures and/or mitigate consequences of failure.
Background notes: .
Maintenance is required to avoid large consequences towards plant availability,
health and safety effects, and environmental aspects of plant operation.
It relates to plant throughput, margins, product quality and off spec products,
toxicity of products, flaring and emissions.
RCM helps to determine the optimal maintenance strategy and preventive

maintenance tasks to provide a safe and reliable process plant.
The outcome can be compared with a maintenance handbook from your car.
This handbook provides the minimum set of tasks to be done periodically to make
sure the car is reliable and safe for your use.
The aim is to get a predictable performance of the equipment so it will be reliable

for operation.

Origin of RCM
S-RCM
Reliability-centred Maintenance: ―a process used
to determine what must be done to ensure that any
physical asset continues to do whatever its users
want it to do in its present operating context‖ * SAE JA1011
1999
Start of efforts by RCM II , John Moubray
commercial airlines to
improve the reliability of
their jets MSG 3 1990
MSG 1
1980
1967
Report: ―Reliability Centered
Maintenance‖ by Nowlan
and Heap for US DoD
* John Moubray, RCM II
Property of Royal Dutch Shell plc RESTRICTED July
July 2010
2010 13
Background notes: .
RCM started off in the airplane industry to reduce the time on the ground
required to keep the airplanes safe and well functioning. A more competitive
maintenance approach was required, moving from time based maintenance to
condition based maintenance.
First recommendations are given by the Maintenance Steering Group 1 in 1968.
RCM is the process to justify and optimize the maintenance activities.

The 7 questions in “classical” RCM are:
• what are the functions and associated performance standards of the asset in
its present operating context?
• in what ways does it fail to fulfill its functions?
• what causes each functional failure?
• what happens when each failure occurs?
• in what way does each failure matter?
• what can be done to predict or prevent each failure?
• what should be done if a suitable proactive task cannot be found?
It is emphasized that Shell-RCM is a streamlined version of RCM and does not

start with a list of all Functions of an asset/equipment; but looks at the functions
related to just the Dominant Failure modes of an equipment.

S-RCM is based on history & experience …
Failures occurred…
• On comparable equipment…
• In comparable operational situations....
Ability to predict or detect degradation

• Is the degradation Age related?
• Is the degradation Revealed?
• Can we detect potential failure?
Understanding the failure

• What causes the degradation
• What influences the degradation?
• What is the consequence of the degradation?
Property of Royal Dutch Shell plc RESTRICTED July

July 2010
2010 15
Background notes: .
Shell RCM is a streamlined version of classical RCM, aimed to drive results faster
by not starting to list out all functions and functional failures.
S-RCM considers failure modes that are most likely to occur – we call these
Dominant Failure Modes.
This excludes aspects like miss-operation and incorrect repairs, issue that might
be analyzed by “classical RCM.
S-RCM may use historical records to identify dominant failure modes and staff
experience to describe the degradation process of the failure mode.
It is important that the more relevant experience the study team has, the better
is the quality of the analysis will be.

The S-RCM process involves a team effort:
Led by the RCM Facilitator, the S-RCM study team comprises of

representatives from:
 Operations
 Process technologist
 Maintenance
 Discipline Engineer (M, RE, I, E)
Information input:
 Technical and operational experience regarding the equipment,
 Maintenance and operations data to feed analysis.
 Design data of the unit and equipment.

Background notes: .
S-RCM analyses aims to get a predictable performance of the equipment so it will
be reliable for operation. As a sound understanding of the Equipment in its actual
operating context is required to perform the RCM analysis, multi disciplinary input
should be provided.
Think through:
• What is your role in the S-RCM study?
• What inputs would be requested from you and how could you prepare?
The basic assumption at the S-RCM analysis is that equipment in a process plant
runs well, in case it is running on design conditions. As such the design conditions
of the plant and equipment are relevant data to be brought to the table by the
participants of the study.
Doing S-RCM studies often drives to better understanding of the participants

related to the reliability of equipment, leading to a cultural change in an
organization with understanding that:
• Maintenance is a process, not a department,
• Maintenance is everybody's business to become reliable,
• Teamwork and cooperation is required to make it work.

S-RCM process: Selection of Equipment.
Selection of
Equipment
Information
Analysis, of Equipment
Feedback &
Review
Task Grouping Dominant

and Execution Failure
Modes
Economic Failure scenario

justification and Criticality
(RtF)
FM characteristics
and tasks
Property of Royal Dutch Shell plc S-RCM - Awareness RESTRICTED July

July 2010
2010 19
Background notes: .
Equipment to be studied in an S-RCM analysis is pre-selected on an estimated
criticality and commonly the higher critical equipment are selected for analysis.
Lower critical equipment can benefit from more generic sets of PM tasks, often
aimed to do no more then reduce consequential damage from a failure.
How do we determine the equipment in our high criticality list?

Selection of Equipment
Selection of equipment for S-RCM analysis

is based on an estimate of the criticality.
— This is the risk determined on a ‗no
maintenance‘ or ‗Run to Failure‘ scenario.
The risk is based on probability and
consequences of failure, regarding:
— Health and Safety (fires, toxic products)
— Economics (production loss, off spec)
HCU recycle gas compressor trip
— Environmental (leakages, emissions)
It is recommended to start at site level and cascade the criticality estimate to
plant level, unit level and equipment level.
Note that equipment level criticality cannot exceed unit level criticality.
This criticality estimate should provide the list of equipment to be analyzed in
S-RCM. In S-RCM the criticality is determined in more detail, so the process
to select equipment should take less time then the RCM analyses itself.
Equipment not selected for analysis should have a RtF criticality.
Background notes: .
The pre-selection of equipment for an S-RCM study seems simple but can be
cumbersome. It requires a failure mode and effect estimate, for which good
plant knowledge should be available.
If the group of equipment chosen to be studied in S-RCM is too large, too much
equipment needs to be studied, causing an overload of the resources involved and
potentially too many tasks specified.
If the group of equipment to be studied in S-RCM is too small, breakdowns might
occur leading to damage and loss.

S-RCM process: Information of Equipment.
Selection of
Equipment
Information
Feedback &
Review

Modes

(RtF)
FM characteristics
and tasks

July 2010
2010 23
Background notes: .
In the S-RCM study the dominant failure modes of equipment are studied.
Information on the equipment is required to get insight in the typical assembly of

the equipment, the degradation mechanisms and the dominant failure-modes
involved.
A first hint on type and assembly of the equipment is often given by the
equipment type, a characteristic of the equipment often available in the plant
Asset Register.

Failure modes
Information of the equipment is required to determine the

degradation processes. This is modelled by Failure Modes.
A Failure Mode = Item + Failure description;

—typically 2 words: a Noun and a Verb.
Failure modes:
Low oil
level
Bearing Lubrication Bearing
failure failure Wear
Oil
degradation
The Item is often a maintainable part of the equipment.

Background notes: .
A Failure Mode is a model of how an Item of an Equipment can fail,
typically provided in 2 words: a part and a failure descriptor.
The part is often maintainable like a bearing, seal or winding.
The descriptor provides insight in the degradation process.
Failure modes can be a combination of multiple aspects that degrade.

A typical failure mode is Bearing Failure, found at analyzing rotating equipment.
The bearing failure is actually bearing wear and the wear can be strongly
influenced by failure of lubrication.
And this can be caused by lack of lubrication medium (Low oil level) or by
degradation of the lubrication medium (Oil Degradation).
Of course the tasks specified to mitigate the risk of bearing failure should look
at oil degradation, oil level and bearing wear all together.
Looking at this we could ask: Should we analyze 3 failure modes or try to do this
in one FM? ( “Bearing Failure”).
When analyzing only one FM instead of the three underlying FM‟s, the underlying
aspects need to be understood nevertheless to provide effective PM‟s.

Determination of Failure Modes.
Engineering
Simple Process
Techninical Diagram
Drawing
2
1
5%
Equipment
10%
Failure Occurrence*
Average (%) of
85%
10% 5%
Mechanical Bearing & Others
seal & seal system housing
FM in a process centrifugal pump
3 Failure
4
Modes
Maintenance History, Incident
data and Experience

Background notes: .
As for S-RCM it is not clear from the start that maintenance tasks
will have economic benefit, first the failure modes are assessed assuming
no maintenance is done, the so called „No Maintenance‟ situation.
Only if the risk of this „No Maintenance‟ situation is not acceptable, a package of
maintenance should be considered.
As the RCM study team should be brought to the same level of knowledge of the
equipment involved, the following information is reviewed:
• An overview drawing of the equipment illustrating typical details subject of
degradation and relevant for reliability, presented by engineering or
maintenance representatives,
• Indication of how the equipment resides in the process flow of the unit and how
it is operated; including equipment around it like instruments with alarms.
This is presented by operations representatives.
• Overview of maintenance history with corrective and preventive maintenance
done, presented by maintenance representatives.
• History on incidents at which the equipment was involved, presented by
operations.
To determine which degradation and failure modes might occur on equipment, it is

recommended to use data available rather than estimates or guesses.

S-RCM process: Dominant Failure Modes.
Selection of
Equipment
Information
Feedback &
Review

Modes

(RtF)
FM characteristics
and tasks
Property of Royal Dutch Shell plc S-RBI level 1 - Awareness RESTRICTED July
July 2010
2010 29
Background notes: .
Shell streamlined RCM focuses on degradations that occur systematically, the so
called Dominant Failure Modes.
In the S-RCM study these dominant failure modes will be studied in detail.
This approach excludes aspects like miss-operation and incorrect repairs on the
equipment. These elements are considered to be covered by knowledge and
competency of people involved, unrelated to preventive maintenance tasks.

Dominant failure modes
100%
Failure
Occurrence
(no maintenance)
55% 35%
10%
Bearing system Seal system Others
FM in a process centrifugal pump
• Pareto concept: Distinguish between the ―vital few‖

(Dominant ones) issues and the ―trivial many‖

Background notes: .
Shell streamlined RCM focuses on degradations that occur systematically in
equipment, the so called Dominant Failure Modes.
Dominant failure modes are failure modes that are:

• Failure Modes encountered on same or similar equipment,
• Failure Modes not yet encountered but considered most likely and credible to
occur (e.g. if existing preventive maintenance would not be executed).
The Dominant Failure modes should cover all degradations that systematically
occur, at least with a probability of once in 20 years or more often.
To list out the dominant failure modes both knowledge and experience of the
actual equipment and of the local situation is required.
In the S-RCM study it is recommended to analyze the Failure Modes in the

sequence of occurrence (in a no maintenance situation), so the fastest
degradation and failure first.

S-RCM process: Failure scenario and Criticality
Selection of
Equipment
Information
Feedback &
Review

Modes

(RtF)
FM characteristics
and tasks

July 2010
2010 33
Background notes: .
When the dominant failure modes are determined, for each failure mode the
degradation and criticality will be determined.
To enable this a failure scenario is to be articulated, describing the failure

development and effects of the equipment and unit.

Failure development: the P-F curve.
1 2
3 4

Background notes: .
Insight in the development of the degradation at a failure mode is
important in an RCM analysis to be able to specify effective tasks.
It involves insight in how the degradation starts, how it develops and how the
functional failure actually occurs.
The minimum acceptable functional condition is relevant and indicated as “the
Failure point”: at that moment a unit can start to lose throughput or products may
become off-spec.
Note that this minimum acceptable condition might happen before the equipment
is actually „Broken Down‟, for instance in case with heat exchanger fouling.
Another important aspect in the degradation is the point where we can detect the
potential failure is coming and this is indicated as “the P point”.
This point is dependant of the degradation mechanism and the method to detect,
typical example is vibration measurement in case of bearing failure of a pump.
Important for the maintenance strategy recommended is insight in the P-F curve
and the time between the “P” and “F” point; to „catch‟ the failure on time.
A longer time between P and F point provides more time to prepare maintenance
activities. The P-F interval determines the interval between condition monitoring
actions to detect the P point and it should be about half the P-F interval in time.

Degradation development for Bearing Failure:
1 2
3 4
Background notes: .
This is an example of a P-F curve for the development of bearing failure on an E-
motor driven centrifugal pump.
Note in RCM we first determine a „ no maintenance‟ (or Run-to-Failure) scenario
as a baseline to verify if the consequences are acceptable in the first place in
case we do nothing.
In this case, the degradation starts by lack of lubrication, this accelerates the
bearing wear, increasing the clearance and causing vibrations : being the P-point.
From the high vibrations the noise increases, bearings get hot, wear accelerates
further and power consumption increases.
In this scenario the E-motor protection detects the required over-power

consumed by the electric motor and switches off the pump.
This degradation description is the first part of the failure mode scenario:
-Low oil level/oil degradation - Bearing wear – Vibration – Noise –
- Bearing gets hot -Increased power consumption - Coupling breaks.
This description helps to articulate effective preventive maintenance tasks.

Failure scenario
A detailed scenario describes the failure of an equipment in a failure

mode. It is used to predict future related issues and consequences of
future failures.
The following information is involved as a minimum:
Function equipment:……. Function item:……. Stand-by:…….
Physical behavior of equipment on failure mode.
Potential for consequential damage to equipment.
How can the failure be observed/detected?
What action will the operator take at the failure:
 To restore production and product quality
 To repair the equipment (priority or scheduled)
What are the consequences (production loss, HSE)
 At repair/ restore: is production loss encountered?
Background notes: .
In the scenario the failure development is described, leading to a stable end
point: the failed state. A scenario is created for each failure mode.
As a failure mode is an item (like bearing) that fails, the function of the item in
relation to the function of the equipment in the plant, must become clear.
In the scenario the consequential damage is indicated, which can be the equipment
itself or connected and downstream equipment.
It is indicated how the failure can be observed or detected – relating to signals
that might come alive in the control room (travelling TOWARDS the operator).
The action of the operator in case of the failure is described in the scenario,
involving the aspects of restoring production and restoring the equipment.
Finally the consequences are provided in terms of economics and HSE.
The failure scenario is articulated by the S-RCM team and the general require-
ment is that it should be credible and factually right. More info can be stored.
The scenario is the core of the S-RCM analysis and should tie the elements like
criticality and preventive maintenance tasks together, in a coherent way.

Cause and Effects of failure modes
System with Filter, Vessel with level controller and a pump.

 Filter fails and Vessels fouls : FM of Filter
 Filter fails and Pump erodes : FM of Filter
 LC fails and pump leaks (by cavitation) : FM of LC
 Oil degrades and pump bearing fails : FM of Pump
Verify if the equipment is

functioning ‗on specification‘: LC
Filter
what failures might happen?
Pump
 What causes the equipment to
function ‗off specification‘ ? Vessel

Background notes: .
Understanding of cause of a failure mode and its effect – with possible conse-
quential/ secondary damage - is relevant to provide efficient preventive tasks.
The tasks should preferably be directed to the cause and not to the consequences.
When a failure mode occurs evidently the item involved fails, for instance a
bearing in a pump. Secondary damage – in case of no maintenance - might be the
shaft, seal and coupling of the pump. Consequential damage might be incurred by
the stopped process flow in other equipment or off spec products produced.
If the FM is attributed to equipment that is not involved in the root cause of the
degradation, ineffective maintenance tasks can easily be specified.
This is illustrated by the example with a vessel, filter, level controller and pump.
Fouling of the vessel can be caused by the failing filter and should not be seen as a
failure mode of the vessel, but of the filter.
In general it is recommended to assess the equipment assuming it is working

in design operating conditions, including start up and shut down conditions .
If equipment is operating outside design conditions degradation might be more
extensive and preventive maintenance can be less effective and reliability might be
low.

Estimated time between failure.
As mentioned in the scenario RCM analysis starts off with a baseline

situation originating from a Run-to-Failure situation (no specific
maintenance executed).
To determine the Risk in case of this Run-to-Failure situation,

the probability of failure is to be estimated.
This probability of failure in the Run-to-Failure situation is called:

Estimated Time Between Failures (ETBF).
The ETBF in an estimated value since for many equipment we have no

experience with a ‗no maintenance situation‘, e.g. in case of an
expensive compressor.
ETBF should be provided by experienced members of the S-RCM

team who have experience on the same or similar equipment and
FM‘s.
Background notes: .
During the S-RCM study the team provides information to articulate the failure
scenario of a failure mode in the „no maintenance‟ scenario (Run to Failure).
Associated to this scenario the time to failure needs to be specified.

This can be an artificial value as for quite some equipment we do not have data for
the situation that no maintenance is executed as it is too costly to try this out
(think of a compressor or gas turbine).
Therefore this timing is called „Estimated Time Between Failures‟, or ETBF.
Together with the consequences the ETBF is used to determine the base risk if
no maintenance is executed.
Failures that occur more often might justify more intense surveillance to the
equipment. Once this surveillance is justified, it might reveal less frequent
failures also. This is the basic reason to start the S-RCM analysis with the
shortest ETBF and to list the ETBF‟s of the failure modes in the beginning of the
study of an equipment.

Criticality from the Risk matrix.
Risk = Probability x Consequence

Probability = Estimated time between failures
RCM ETBF
Criticality
H 0.0 - 0.5 y L MH H E E
PROBABILITY
M 0.5 -4.0 y L M MH H E
L 4.0 - 20 y N L M MH H
N > 20 y N N L M MH
no/slight damage minor damage local damage major damage extensive

ECONOMICS (US$)
CONSEQUENCE
(<10k) (10-100k) (0.1-1m) (1-10m) damage (>10m)

CATEGORY
HEALTH & SAFETY no/slight injury minor injury major injury single fatality multiple fatalities
ENVIRONMENT no/slight effect minor effect local effect major effect massive effect
CONSEQUENCE CLASS NEGLIGIBLE LOW MEDIUM HIGH EXTREME

Background notes: .
The Risk Assessment Matrix (RAM) is used to determine the Criticality of the
Failure mode. The RAM in RRM has 6 risk levels:
Negligible, Low, Medium, Medium-High, High and Extreme.
The Criticality of a Failure Mode is the Risk of failure in the „no maintenance‟
situation, the criticality is also called Base Risk.
The criticality is determined by the Probability of failure and the Consequence of
failure.
• The Consequence of failure comes forward from the information provided in
the scenario for Economics, Environment, Health and Safety.
• The Probability of failure comes forward from the Estimated time between
failure.
Information of the plant is required to determine the economical consequences in

case of lost production or off spec production.
The monetary value of an hour lost production is used in the determination of

economic consequences, and is called the „Product Loss Equation‟.

Example P301A, failure scenario for Bearing Failure:
Function equipment: Feed unit with 15 bar 300 m3/h.

Function item: (roller bearing) Keep shaft in place.
Stand-by: A is main pump and B is Stand-by (Man. Start).
Physical behavior of equipment on failure mode.
Low oil level/oil degradation, bearing wear, vibration, noise,
bearing gets hot, increased power consumption, coupling breaks.
Potential for consequential damage to equipment.
Bearings, shaft, seal, coupling.
How can the failure be observed?
Flow meter of furnace shuts down the furnace and gives alarm.
What action will the operator take at the failure:
Switch to B pump. Scheduled repair.
What are the consequences (production loss, HSE)
3 hours production loss when restarting the furnace.
No Health and Safety aspects. Slight leakage in plant.

Background notes: .
Bearing failure on Centrifugal pump P301A is a calculation example.
This is an example of a scenario for the failure mode Bearing Failure on a 180 kW
feed pump. During the S-RCM study the team provides information about the
pump: bearing type, seal type, number of impellers, etc.
The stable end point of the Failure Mode is that the coupling breaks caused by
the overpower dissipated by the damaged bearings. As the coupling breaks the E-
motor is still running (not causing alarm).
The low flow will cause a shut down of the furnace by the instrumented protective
system, protecting the furnace and causing an alarm in the control room.
The operator can restore production by starting the stand by pump (manually)
and restart the furnace.
To restore the equipment a scheduler repair is initiated .

Finally the consequences are provided in terms of economics and HSE.
It is given that when this unit fails, the costs of lost production are 2000 $/hour.

Estimated time between failure P301 A Bearing Failure.
The information in the scenario should provide background for the

Estimated Time Between Failure in the Run to Failure scenario.
In this case the bearing wear is accelerated by the lack of
lubrication.
The Estimated Time Between Failures (ETBF) is 2 years,
based on the fact that oil is not checked or exchanged.

Background notes: .
Together with the consequences the ETBF is used to determine the base risk if
no maintenance is executed.
The ETBF is the probability of failure and used to determine the Base Risk of the
failure mode.
Note that relative low ETBF‟s might indicate design problems of the equipment.
Equipment in a situation with an inherent low reliability can not made reliable by
providing more preventive maintenance. In such a situation one should go through
„Root Cause Analysis‟ and improve the inherent reliability.

Consequence of failure P301 A Bearing Failure
Questionnaires for:
Economic:
3 hours production loss when restarting the furnace: 6000 $.
Plus: bearings, shaft, seal, coupling repair: 15,600 $
 11,000 $ materials; 1000 $ crane; 80 h labor @ 45$/h.
Health & Safety: Nothing
Environmental: Slight effect (3200 $)
Total consequences = 24,800 $ (Price of Non Conformance)

Background notes: .
The consequences of failure for economics involve production loss, costs of off
spec production and repair costs. It is given that when this unit fails, production
loss costs $ 2000 /hour.
The production loss costs is then $ 6000

The repair costs is 11,000 $ materials; $ 1000 crane; 80 man hours labor @
$ 45/h: total $ 15,600.
Total economic loss is $ 21,600.
On the Environmental consequence is “slight”: a financial value of $ 3200 can be

granted.
The total consequences in financial value are about $ 24,800 in the „no
maintenance‟ case. This is called the „Price Of Non Conformance‟ in RCM (PONC).
In general during the S-RCM analysis it is assumed the spare parts required to
repair the dominant failure modes are available.

Criticality for P301 A Bearing Failure is Medium level.
RCM ETBF
Criticality
H 0.0 - 0.5 y L MH H E E
PROBABILITY
M 0.5 -4.0 y L M MH H E
L 4.0 - 20 y N L M MH H
N > 20 y N N L M MH
no/slight damage minor damage local damage major damage extensive

ECONOMICS (US$)
CONSEQUENCE
(<10k) (10-100k) (0.1-1m) (1-10m) damage (>10m)

CATEGORY
HEALTH & SAFETY no/slight injury minor injury major injury single fatality multiple fatalities
ENVIRONMENT no/slight effect minor effect local effect major effect massive effect
CONSEQUENCE CLASS NEGLIGIBLE LOW MEDIUM HIGH EXTREME
Average cost of this risk = $12,400/year.

Background notes: .
The data in the Risk Matrix indicates the criticality for P301 A Bearing Failure is
Medium level.
The total consequences in the „No maintenance‟ case has a financial value of
$ 24,800. This is called the „Price Of Non Conformance‟ in RCM (PONC).
The Estimated Time Between Failure is 2 years
On average the risk of this failure mode costs $ 12,400 /year.

Estimated Time Between Failures: no maintenance done
Example: Pump bearing fails every 2 years if no maintenance is done.

ETBF:- 2 years
Condition
0 1.5 year 2 year Time
Install new pump Bearing failure

and START Seal Failure
Shaft and
Oil degradation/low level Coupling
Increased vibration Damage,
Production loss
Property of Royal Dutch Shell plc
PONC= $ 24,8k RESTRICTED July 2010 55
Background notes: .
Illustration of the degradation in time for Bearing Failure in the
„No maintenance‟ scenario, also called Run to Failure scenario:
• Low oil level/oil degradation,
• Bearing wear,
• Vibration,
• Noise,
• Bearing gets hot,
• Increased power consumption,
• Coupling breaks.

Criticality results
The criticality levels are used to justify preventive maintenance tasks.

In general in process plants (E) Extreme level criticalities are rare:
risking these consequences in case of ‗no maintenance‘ is often
‗designed out‘ by for instance safeguarding systems.
As S-RCM should focus on higher levels of criticality, one might want
to spend less time
on (N) Negligible
or (L) Low
level criticalities.

Background notes: .
One might now appreciate the assessment of the „no maintenance‟ scenario in RCM.
The higher the criticality level found, the better the maintenance strategy should
be managed. However, if the criticality level is low, the Run to Failure scenario
might just be acceptable.
In general it is recommended to direct the time spend in the RCM analysis

towards higher criticality levels (H, HM and M) and try to keep the number of
these analyses above 1/3 of all RCM analysis done.
The distribution of criticalities found can be used to verify the pre-selection of

the equipment for S-RCM analysis in the first place. This should lead to improved
selection in next S-RCM studies.
In general, in well designed process plants, Extreme level criticalities are rare as
risking these consequences in case „maintenance is not done‟ is too risky.
When Extreme criticality levels are found, one might want to review the analysis
as there might be instrumented protective functions available which are not fully
taken into account in the failure scenario.

Criticality results
From an economical point of view the amount of risk is providing

justification for the preventive maintenance tasks recommended.
If the risk is low, the preventive maintenance efforts may not be worth
doing.
During an S-RCM session the team members with technical
background often find it difficult to articulate equipment degradation
till the point it breaks down and fails.
This might lead to an underestimation of the true criticality, denial of
the real risks and false rejection of potential preventive maintenance
efforts. As such we should note:
More Risk justifies More Preventive Maintenance

Also: No risk = No justification for Preventive Maintenance.
Background notes: .
It is important to make a realistic estimate of the risks involved, both too high as
too low estimated risk levels should be avoided.
If the risk is low in the „no maintenance‟ scenario then there is no justification to
do preventive maintenance activities in the first place.
One should keep in mind that more risk in general will help justification to do more
preventive maintenance efforts.
Sites with a strong fire fighting mentality often tend – in first instance - to
underestimate the consequences in a „no-maintenance‟ scenario. Also often
technical people find it difficult to describe how equipment will degrade, fail and
break down.
Detailed insight in the degradation and break down is however important to

determine if maintenance is required in the first place, and how maintenance
tasks can effectively assist in mitigating the risk of failure.
At determining the Base Risk in the S-RCM of a failure mode, a significant step is
completed and next attention is drawn towards: What to do about it?

S-RCM process: FM Characteristics and tasks.
Selection of
Equipment
Information
Feedback &
Review

Modes

(RtF)
FM characteristics
and tasks
Property of Royal Dutch Shell plc S-RCM - Awareness RESTRICTED July 2010 61
Background notes: .
In S-RCM maintenance tasks are optimized to provide a cost effective
maintenance strategy for that particular failure mode.
To determine this, firstly one should find tasks that can effectively preserve the
functions, prevent premature failures and/or mitigate consequences of failure.
These tasks are dependant of the characteristics of the failure mode involved.
The main characteristics are:
• Is the failure age related?
• Is the failure hidden?
These characteristics can lead to:

1. Scheduled maintenance (or time based maintenance) ,
2. Condition based maintenance (with condition monitoring).
After the tasks are articulated an economical verification is done to determine

cost effectiveness of the maintenance strategy suggested.

Maintenance options:
Property of Royal Dutch Shell plc S-RCM Awareness RESTRICTED July 2010 63
Background notes: .
Starting from an „Un managed‟ maintenance approach, the „Managed‟ maintenance
approach (1-2) one looks first to preventive maintenance, what actually can break
down into two strategies:
• Condition based maintenance, using condition monitoring tasks to decide what
the optimal moment is to execute the maintenance,
• Scheduled maintenance, with tasks executed time based; independently of the
state of the equipment.
The maintenance strategy chosen is dependant of the failure mode characteris-
tics and the circumstances.
Based on the economics of the tasks and consequences involved, one might even
decide to do no preventive maintenance (3):
• No preventive maintenance: decide to a „Run to Failure‟ scenario as there is no
cost effective preventive maintenance package.
To drive from an „Un managed‟ maintenance approach towards the „Managed‟

approach the S-RCM is an excellent methodology to use.

Target Area for Preventive Maintenance Strategies:
Background notes: .
When S-RCM is used to optimize the maintenance strategies, we move away from
the unmanaged, reactive approach.
S-RCM focuses on the higher criticalities and all breakdowns with negligible
consequences are given a Run-to-failure maintenance strategy.
Based on the failure mode characteristics and the circumstances, Condition based
maintenance or Time based (Scheduled) maintenance is applied.

Characteristics of a failure mode: Age Related
For an economic set of preventive maintenance tasks the

characteristics of the failure mode should be known.
A Failure mode is Age related if the degradation is mainly dependent

of time.
 An Age related failure mode can be maintained with a time based
task.
A Failure mode is Non-age related if the degradation is dependant of
several factors, like:
 Time, Pressure
 Load, Vibrations
 Speed, Temperature.
Background notes: .
For an economic set of preventive maintenance tasks the characteristics of the
failure mode should be known.
There are two typical FM characteristics:
• Age related and non-age related,
• Revealed and Un-revealed.
Age related failure modes can be maintained by time based maintenance.

Non-age related failure modes can be maintained by condition based maintenance.
Note that even if a failure mode is non age related it still can have a clear P-F
curve, useful to determine degradation has started and failure is going to happen.
This provides predictability and enables precautions be taken before the failure
actually occurs, potentially avoiding consequential damage

Example of age and non-age related failure pattern.
Age Related Non-age related
P-F curves P-F curves

Condition
of items
of items
ConditionNo. of Failures
No. of Failures
Conditional Prob, %
Failure Rate or
Conditional Prob, %
Failure Rate or
Failure pattern
Failure pattern
Useful life
Time Time
Background notes: .
Age related failures show a strong dependency in time, while other factors have
minor to no influence on the degradation. Age related failures can effectively be
maintained with time based maintenance (also called scheduled maintenance).
Non Age related failures show a dependency of multiple factors. This can be time,
but also load, pressure, temperature and e.g. vibrations. Non Age related failures
can effectively be maintained with condition based maintenance.
Non age related failures without a P-F curve can be fully random failures,
comparable with how electronics fail.
To illustrate the difference between Age related failures and Non age related
failures for both graphs are shown with:
• P-F curves for multiple comparable types of equipment,
• A distribution of the failures in time,
• The Conditional Probability of Failure: often referred to as „the bathtub
curves‟.

Characteristics of a failure mode: Un-revealed failures.
Some functions can be hidden:

Under normal circumstances the function is not evident and a separate
(unrelated) event is required to make use of it.
Examples are:
 A stand-by pump
 An over-pressure protection valve
 An (electrical) over-power protection
A failure on such a function will also not be evident, till the separate
(unrelated) event occurs, this is called a hidden failure or unrevealed
failure.
To reveal if a hidden function is lost and an unrevealed failure has
materialized, action is required to test or restore the function
periodically.
Background notes: .
A function is hidden when under normal circumstances the function is not evident
to the operating crew on it own. Only with a second – independent event – the
function will be used and a failure of this function can be revealed.
Such a failure is called an unrevealed failure and active testing is required to find
such a failure.
The test frequency can be determined based on two aspects:

• The mean time between failure of the unrevealed failure mode and
• The required „probability of failure on demand „of the hidden function.
If no tasks are put in place to determine the unrevealed failures, the failures can
cause surprises in daily operation of a plant, potentially leading to high priority
maintenance requests. As such it is recommended to double check if a Run to
Failure maintenance strategy is considered on an unrevealed failure mode.
The bearing failure of P301A is non-age related and revealed. In next graphs the
effect of a set of preventive maintenance tasks are shown.

Suggestion of Preventive Maintenance Tasks (P301A)
1 2
3 4

Background notes: .
For the Bearing Failure of P301A in the „No maintenance‟ scenario, the following
physical behavior was described:
• Low oil level/oil degradation,
• Bearing wear,
• Vibration,
• Noise,
• Bearing gets hot,
• Increased power consumption,
• Coupling breaks.
This physical behavior can now be used to specify effective tasks:

• Check bearing temperature (condition monitoring),
• Check oil level (condition monitoring),
• Vibration measurement and trending (condition monitoring),
• Replace oil time based,
• Condition based repair of bearing – before consequential damage occurs.
This will elongate the time to failure and detect potential failure as shown in
steps (1)-(2)-(3)-(4).

S-RCM process: Economic justification
Selection of
Equipment
Information
Feedback &
Review

Modes

(RtF)
FM characteristics
and tasks
Background notes: .
Next step is to determine if the suggested maintenance strategy can reduce the
risk and if the maintenance effort involved results in an economic profitable
position.
In the example of the bearing failure mentioned before, the base risk (M level
criticality) involved an average monetary value of $ 12,400 /year.
This is the financial base to do the maintenance package.
In the following slides the economic verification of a package of preventive

maintenance tasks is explained.

Cost optimization in S-RCM
Determine preventive maintenance tasks leading to cost optimization in

S-RCM.
1 2
Background notes: .
In the previous steps preventive maintenance tasks were suggested for the
failure mode: Bearing Failure. Looking at maintenance on equipment, there are 3
area‟s to identify:
• Not enough maintenance done: under-maintained.
• Too much maintenance done: over-maintained
• Balanced position, the optimum area.
The graph shows an area of optimum maintenance efforts (not just one point).
The objective is to relate the maintenance efforts to effective mitigation of the
risks of the failure mode. It boils down to the question: what maintenance
efforts do effectively contribute to a reduction or mitigation of the risk.
These efforts should preserve the functions, prevent premature failures and/or
mitigate consequences of failure.

Effectiveness of in Tasks (P301A)
The effectiveness that the Run-to-Failure
consequences might not happen again is
expressed in the ETBC, here once in 18 years.
Condition
0 Time Missed! 18 years

Oil
top up
Vib.
Meas.
$ Bearing failure
Condition Condition Seal Failure
based based Coupling breaks
Install new pump Bearing Bearing Shaft damage
and START replacement replacement
Production loss
$$ $$
PONC=$24,800

Background notes: .
A maintenance strategy might fail after all by reasons that signals are missed
(high temperature, noise, vibration) or tasks are skipped, and as such the „Run to
failure‟ scenario might happen in the end.
In the S-RCM methodology this is anticipated as a residual risk of the

maintenance strategy chosen. This residual risk is introduced as a probability of
„Estimated Time Between Consequences‟ (ETBC); here 18 years.
This residual risk should be taken into account at determining if a maintenance
strategy is cost effective, and is added to the „do maintenance‟ costs.
The ETBC leads to a residual risk level (relative to the base risk – or criticality
level) and the maintenance strategy should result in sufficient risk reduction.
S-RCM studies in Shell the residual risk level should be below MH level.
Actually the ETBC value is a confidence rating that the package of PM tasks be
able to mitigate the risk of failure. A weak set of tasks might result in a shorter
ETBC value.
An ETBC of 18 years brings the residual risk to an acceptable „Low‟ level.

Maintenance Efficiency Index: MEI
MEI =
(Base Risk) - (Residual Risk)
(cost of the maintenance plan)
S PONC
ETBF - S PONC
ET BC
MEI =
Annual POC
MEI must be > 1 to approve tasks !

Rijswijk – The Netherlands S-RCM
S-RCM AwarenessAwareness
Background notes: .
Although the comparison „No maintenance – Do maintenance‟ can show the
economically strongest alternative, it does not enable to determine the most cost
effective maintenance alternative.
For this the Maintenance Efficiency Index (MEI) is introduced, where the effect
of the task in risk reduction is compared with the cost of maintenance package.
The risk reduction is the difference of the base risk and the residual risk.
The MEI is the ratio of the risk reduction and the maintenance costs.
This is also a typical S-RCM feature that helps identifying economically strong
maintenance packages
If the MEI equals to 1 then the benefits in risk reduction equal the PM costs
and no financial benefit is gained.

Maintenance Efficiency Index: MEI.
Although the comparison ‗No maintenance – Do maintenance‘ can show

the economically strongest alternative, it does not enable to determine the
most cost effective maintenance alternative.
For this the Maintenance Efficiency Index (MEI) is used in S-RCM.
The reduction of risk due to the maintenance package is
( using the figures of the very first example in the course, remember? ):
Criticality - Residual Risk 50 k$/y – 20 k$/y

Maintenance Costs 40 k$ /y
MEI = ¾ = 0.75
An MEI of 1 indicates that the potential savings - of the reduced risk - in
monetary terms equals the maintenance costs.
Or: the maintenance effort does not earn money.
For a sound maintenance strategy the MEI should preferably be > 2
Background notes: .
If the MEI is below 1 than the preventive maintenance package not economically
interesting, and a run-to-failure scenario is recommended (resulting in an MEI of
0).
MEI‟s scores between 1 and 2 often indicate marginally beneficial preventive

maintenance packages, an indication of being ineffective to deal with the cause of
the failure and dealing with the consequence of the failure.
It is recommended always to strive to the highest MEI score.
It is shown that not all maintenance activities might be cost effective and that
this can be determined by calculating the MEI score.
To come forward with a cost effective package of preventive maintenance tasks it

is important to specify tasks that are effective to
1. Preserve the functions,
2. Prevent premature failures and/or
3. Mitigate consequences of failure.
Maintenance task that are directed to the consequences have generally lower MEI
scores, compared to tasks directed to the causes.

Preventive Maintenance package for P301A:
The tasks involved are:

 Operator checks on round to check oil, bearing temperature and noise:
every day 2 minutes costs: $ 852/yr
 Vibration monitoring every month for half an hour, costs $ 300/yr
 Schedule oil change for half an hour every year: costs $ 35/yr
 Condition based replacement of bearings and seals
every 3 years for one week, costs 8400 $, or: $ 2800/yr.
So this total maintenance package would cost: $ 3987/year
The ETBC=18 years so the residual risk is 24,800 $/18yr = $ 1378/y
So: The No-maintenance situation cost: $ 12,400/yr
The Do-maintenance situation cost 3987+1378 = $ 5364/yr
The MEI is (13,400-1378)/5364 = 2.05
Background notes: .
The task frequencies for condition monitoring should be based on the P-F curves
of the underlying degradation mechanisms.
From the suggested set of preventive maintenance tasks the yearly costs are
determined, resulting in the cost for the preventive maintenance package.
In this situation four tasks are suggested with a total cost of $ 3987 /yr.
The rest risk in monetary terms is $ 1378/yr, what should be contributed to the
„do maintenance „ costs.
The „No Maintenance‟ situation costs $ 12,400/y and the „Do Maintenance‟
situation costs $ 5364/y: so its more beneficial to do the maintenance.
Based on Risk reduction and maintenance costs, an MEI is determined of 2.05;

indicating a cost effective maintenance package.
A rule of thump for the yearly maintenance cost is that it should be about 1/3 of
the PONC ($/year).

Economically approving tasks
MEI=0 0 < MEI < 1 1 < MEI < 2 MEI > 2
No cost efficient
PM tasks:
Improve
Marginally
maintenance
justified PM tasks.
strategy or apply
Run to Failure Verify Data on
Accuracy Stronger maintenance
strategy.
Avoid fall back at
Run-to-Failure next review. Approved tasks
Strategy for FM
No PM tasks …continuously strive to highest MEI as this
leads to most effective maintenance efforts!
Background notes: .
For an economical justification of maintenance tasks we basically see four area‟s:
• MEI > 2: indicating economically justified maintenance strategies.
• 1< MEI <2: indicating the maintenance strategy marginally economic justification.
These analyses can be accepted but could indicate the true cause of the
degradation and failure is not know in detail.
This might trigger collection of additional operational and technical history at
next maintenance actions. Next the options are:
• If there is credit that the strategy is stable, do the PM's,
• In doubt, execute the PM's with remark to collect field data and do a short
notice review,
• In big doubt, choose a RtF strategy and collect data to reconsider later.
• 0< MEI <1: indicating the maintenance strategy is not justified on economical
basics. This might trigger investigation of more effective or efficient tasks and
strategies. This MEI will recommend no preventive maintenance being executed:
Positively choose for a Run to failure strategy. The MEI will be 0.
• MEI = 0: Indicates a Run to failure strategy is selected: no risk reduction.

Steps during S-RCM assessment for a tag
5%
10%
Function?
Failure Occurrence*
Average (%) of
Consequence
85%
of failure?
10%
Mechanical Bearing &

5%
Others
How discovered?
seal & seal system housing
FM in a process centrifugal pump What action?
Production loss?
Select Select HSE?

Define Assess
Tag FM Scenario Criticality
‗No Maintenance‘
‗Do Maintenance‘
Tasks; Freq;
ETBC
FM characteristics WC; POC;
MEI
Ex/New/Mod
Initial Risk Maintenance Residual

Cost + Risk

Background notes: .
Summary of the particular analysis steps to do S-RCM analysis with the IT tool
during a workshop. It shows the sequence how the information that is collected in
the process steps until the failure mode is analyzed.
It starts with the „No Maintenance‟ scenario leading to the equipment criticality
for that failure mode (in a Run to Failure scenario).
From this base line a preventive maintenance package of tasks is determined

based on the characteristics of the Failure Mode. However, if the risk is
negligible one might opt for a Run to Failure strategy directly at this point.
Based on the effectiveness of the preventive maintenance package to mitigate

the risk, the ETBC is estimated leading to the residual risk level, and
determination of the MEI score.
Note: the example with P301A involved one FM and resulted in 4 tasks.
Normally in an S-RCM study, up to 200 equipment can be analyzed in a process
unit, with 350 FM‟s and 650 tasks. Though theoretically the loop shown above
should be circled 350 times, in practice many parts can be copied alternatively.

S-RCM process: Task grouping and Execution
Selection of
Equipment
Information
Feedback &
Review

Modes
Economic Faiurel scenario

(RtF)
FM characteristics
and tasks
Background notes: .
As mentioned: RCM is a process to specify technically appropriate, feasible and
economically justified actions to maintain equipment.
Next step is to hand over the tasks and execute them in the field.

Handover to execution: focus on efficiency
Objective: CM tasks
Technically S-RCM process: &
appropriate, Justification of
feasible and
TB tasks:
preventive
economically maintenance tasks Ability to
justified
forecast
maintenance tasks.
execution.
Operator tasks,
Maintenance Execution
Objective: Prepare
Maintenance and/or Turn Around.
Maintenance
history Efficient execution of activities for most
tasks. effective execution.
Background notes: .
The output of S-RCM is a set of tasks for various equipment and failure modes.
For scheduled execution the Time based tasks and Condition monitoring tasks are
handed over. These tasks can flow into the processes of: Daily Maintenance,
Operator task management (operator rounds) and to Turn Around tasks.
A set of preventive maintenance tasks has the unique benefit that the work can
be forecasted well in advance. To materialize this benefit the preventive tasks
should be grouped into efficient work packages and translated into automatically
called work requests ready for execution in the field.
When receiving the (pre-grouped) preventive maintenance tasks, the maintenance

(preventive) planner will do the next steps:
• Verify if additional information is needed for the tasks, like: procedures, work
instructions, registration forms, materials (type of oil) and/or spare parts.
• Verify if extra tasks are needed like: insulation, scaffolding, crane, transport.
• Decide if a group of task can be executed in one or multiple work orders.
• Decide when the task should be executed next (at long task intervals ).
• Create the objects in the Maintenance Management System to automatically
create the preventive work orders in time.

Task grouping for efficient execution
Grouping of tasks provide efficiency at execution by reducing losses from
preparation work, travel time and closeout activities.
Tasks can be grouped for instance, if they require comparable:
Tools (crane), work procedures and instructions,
Preparation, drawings, history capturing,
Type of repairs and/or spare parts (activities).
Craft and/or contractors (e.g. contract for labor),
Start-up or shut-down activities (same unit/system),
Travel actions (Work locations),
Operational clearance (Permits).
Examples:
 Test runs of stand-by pumps combined with oil change, instrument checks
and electrical checks.
 Oil changes combined with regular plant stops.
Background notes: .
Task details relevant for execution are important to group tasks in efficient work
packages.
The S-RCM Facilitator – in charge of executing the S-RCM study – and the
Maintenance (Preventive) Planner – in charge of entering the tasks in the system –
have a shared responsibility to make sure all the tasks from the study are handed
over to execution in a timely manner.
When tasks are not grouped in the system for execution, a highly scattered
series of tasks could be directed towards maintenance planning and scheduling.
There, the tasks might be combined by hand by the maintenance scheduler after
all, at each execution again. This might illustrate that the task grouping
preferably should be done before the tasks are entered in the system.
It is recommended to start this task grouping at the end of the S-RCM study.
For a typical S-RCM study with 600 tasks, it is much more efficient to hand over
25-30 pre-grouped sets of tasks to the maintenance (preventive) planner, instead
of the list of 600 tasks.
Although the explanation above is articulated in terms of task handover to
maintenance, comparable activities could take place for handover of operator
tasks and Turn Around tasks.

S-RCM process: Analysis, Feedback and Review.
Selection of
Equipment
Information
Feedback &
Review

Modes

(RtF)
FM characteristics
and tasks
Background notes: .
As it is not possible to describe exactly what a good S-RCM study is
(only what a bad S-0RCM study is), it is important to apply
learning moments in the S-RCM process to enable improvement.
Here the classical Plan-Do-Check-Act loop should be used.

Living program to continuously improve S-RCM
Feedback in Living program:
S-RCM process:
Reactive feedback:
• Failures from maintenance Justification of
requests, preventive maintenance
• Consequence of failures,
• Time between failures.
tasks
Pro active feedback:

• Ineffective PM tasks,
• Inefficiently grouped
tasks,
• Technically unfeasible
tasks. Operator tasks,
Maintenance Execution
Data updates: and/or Turn Around.
• Plant changes, projects. Efficient execution of tasks.
Background notes: .
When maintenance is executed there can be good learning-points to continuously
improve the S-RCM analysis.
Here both preventive and corrective maintenance is of interest.
Aspects of interest from field maintenance execution for S-RCM analyses are:
• Evidence of failure modes ( what parts are damaged ? ),
• Time between failures ( shorter than expected ? ),
• Consequence of failures ( priorities differ much ?),
• Cost of maintenance. ( preventive and corrective tasks ?)
• Feedback from field execution should be used to periodically update the S-
RCM study.
Note that projects and plant-changes should be assessed on relevance to update

the S-RCM study of the unit involved.
Specifically when equipment is removed the PM tasks shall be deleted.

S-RCM – Feedback loop
S-RCM
Equipment
Data List Initial S-RCM Investigate/Redesign
Analysis Recommendations
Historic Data
CM, PM, Root cause analysis
Incidents
PM Tasks
List
Corrective
feedback. Feedback Reliability Maint. Mgt RM Job
System Requests
Proactive
feedback. Optimise
Data Costs Maintenance
updates. Execution
RM & PM Jobs
Analysis
Availability Plan &
Schedule Jobs
Execute Jobs
PM Compliance Ops & Maint.

Background notes: .
This overview illustrates the position of the S-RCM process relative to the
Maintenance execution process and the process of Root Cause Analysis/Defect
Elimination.
The S-RCM process specifies preventive maintenance tasks to be executed.

In case the S-RCM identifies extensive degradations, extreme criticalities or low
scores in MEI, a Root Cause Analysis can be used to identify the background and
remedial actions.
Note that many incidents are followed up with investigations where the causes are
reported including „failure mode and effect analyses‟. Often these analysis provide
most of the information needed to perform an S-RCM analysis and enable
preventive maintenance tasks to put into place.

S-RCM – Typical timeline in implementation
Training 1 week
2 - 6 weeks
Data Collection
3 - 7 weeks
S-RCM study
6 - 20 weeks
• Terms of Reference pre Implementation

• Maintenance History
• Workshop schedule
• Asset register Continually
PM Compliance
• Analyze Failure Modes
• Specify tasks
• RCM methodology • Track progress
• RCM study approval
• IT tool usage • Register improvements Timely execution
• Implementation timeline
• Workshop preparation • End report + Q check and completion
• Track progress in Maint.
• Workshop organization • Task grouping of PM tasks.
Management System.
Property of Royal Dutch Shell plc

Rijswijk – The Netherlands RCM facilitator is key! RESTRICTED July 2010 103
Background notes:
The training session before the S-RCM study is also used to frame out the timing
and preparation work for the S-RCM study to be done.
The RCM facilitator has a key role to prepare, execute and complete the RCM
study, including handover to execution.
Preparation involves selection of the Tag‟s to be studied, and plan the time needed
for the study – preferably grouped per discipline.
The participants in the S-RCM study will be approached early before the
workshop(s) to be able to prepare, gather data and commit the timing in their
schedules.

THANK YOU !
You completed the S-RCM awarenes
training
Now please proceed to the quiz to continue and obtain your
certificate.

Background notes:
Congratulations, you completed this course, we hope you enjoyed it!
Note on the first slides we asked you some questions regarding your new job on a
plant with 2000 doubtful preventive maintenance tasks in place.
Hopefully you understand now how to verify these tasks, no?

We would recommend to make sure:
• The tasks address credible failure modes,
• The consequence of failure can (not) be accepted,
• The maintenance strategies applied do reduce risk and have an MEI >2.
• The preventive maintenance tasks are executed on time,
• Feedback from task execution is captured and used to improve,
• High priority and costly maintenance requests still generated are analyzed.
Shell Reliability Centered Maintenance is the method used for this.
Thanks for conducting the training and success with completing the quiz!


S-RCM Awareness Training P&T R7a 23jul2010

Uploaded by

Copyright:

Available Formats

S-RCM Awareness Training P&T R7a 23jul2010

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

S-RCM Awareness Training P&T R7a 23jul2010

Uploaded by

Copyright:

Available Formats

What is Shell Reliability Centered Maintenance (S-RCM)?

What is Shell Reliability Centered Maintenance (S-RCM)?

What are the steps involved in conducting an S-RCM study?

What are the steps involved in conducting an S-RCM study?

SHELL RELIABILITY CENTERED MAINTENANCE

COURSE, LEVEL 1 - AWARENESS

Introduction into the concepts

Frat Amarra, Nagapati Hegde and Harry van Teijlingen

Property of Royal Dutch Shell plc RESTRICTED July 2010 1

This web based training takes about 3 hours to complete.

Property of Royal Dutch Shell plc RESTRICTED July 2010 2

Please take a sheet of paper and a pen, ok?

Imagine you just started in your job on a plant:

How would you do this?

Property of Royal Dutch Shell plc RESTRICTED July 2010 3

However the same failure of an equipment can have high consequences if it is on a

This analyses are done in Shell Reliability Centered Maintenance.

Property of Royal Dutch Shell plc RESTRICTED July 2010 4

If the pump fails the consequences will be a loss of $ 100k.

A maintenance plan is recommended and costs $ 40k per year.

Now: Do you think is this a good preventive maintenance plan? Y or N

Property of Royal Dutch Shell plc RESTRICTED July 2010 5

Please calculate a cost effective approach: To execute the tasks or not?

It is recommended to choose a time frame, e.g. 1 year

And when we do executing the preventive maintenance tasks,

What is your choice?

Property of Royal Dutch Shell plc RESTRICTED July 2010 6

If the pump fails the consequences will be a loss of: $ 100k.

The maintenance plan costs $ 40k per year.

Together the cost in the ‗maintenance‗ situation would be:

Property of Royal Dutch Shell plc RESTRICTED July 2010 7

However the costs of the maintenance tasks is relatively high,

The verification of the cost effectiveness is part of Shell Reliability Centered

S-RCM helps to select cost effective preventive maintenance tasks and to

In this case it could lead to two next steps:

Property of Royal Dutch Shell plc RESTRICTED July 2010 8

Property of Royal Dutch Shell plc RESTRICTED July 2010 9

S-RCM analyses equipment failures, looks at the consequences of loosing

S-RCM is one of the elements of reliability management of a process plant.

Property of Royal Dutch Shell plc RESTRICTED July 2010 10

RCM is about maintenance activities.

RCM is a process to specify technically appropriate, technically

RCM helps to determine the optimal maintenance strategy and preventive

The aim is to get a predictable performance of the equipment so it will be reliable

Property of Royal Dutch Shell plc RESTRICTED July 2010 12

RCM is the process to justify and optimize the maintenance activities.

It is emphasized that Shell-RCM is a streamlined version of RCM and does not

Property of Royal Dutch Shell plc RESTRICTED July 2010 14

Ability to predict or detect degradation

Understanding the failure

Property of Royal Dutch Shell plc RESTRICTED July

Property of Royal Dutch Shell plc RESTRICTED July 2010 16

Led by the RCM Facilitator, the S-RCM study team comprises of

Property of Royal Dutch Shell plc RESTRICTED July 2010 17

Doing S-RCM studies often drives to better understanding of the participants

Property of Royal Dutch Shell plc RESTRICTED July 2010 18

Task Grouping Dominant

Economic Failure scenario

Property of Royal Dutch Shell plc S-RCM - Awareness RESTRICTED July