Maintenance
&
Reliability
Overview
Presenter
Charles Marks
Reliability Engineer with over 40
years of industrial maintenance/
reliability experience in oil and gas:
metals processing, manufacturing,
and mining.
Director at Fluor and has
done multiple projects for BP.
2
M
Expectations
• Provide overview of maintenance and reliability concepts
• Introduce maintenance and reliability concepts that oil, gas, and
petrochemical employees may not have encountered
• It is impossible to cover all maintenance and reliability concepts
in this brief discussion
3
C
History of CMRP and BoK
• The Society of Maintenance and Reliability Professionals,
SMRP, developed a certifying organization, SMRP Certifying
Organization, SMRPCO circa 1997
• SMRPCO developed the Certified Maintenance and Reliability
Professional certification program, (CMRP)
4
C
Thoughts on Certification
• No pre-qualifications required to take the test
• One can not easily study for the test
• The more experience one has, the better chance one has to
pass the test
• About 64% of all people that take the CMRP test pass.
• The test can be retaken every 6 months
5
C
Body of Knowledge (BoK)
• The BoK captures key maintenance and reliability principles
• The presenters have reviewed the BoK and have selected key
elements to discuss
• Not all items will be discussed
• The BoK can be obtained from SMRP at www.smrp.org
• If you care to take the CMRP test the CMRP Candidate Guide
for Certification and Re-Certification, with sample CMRP test
questions, can be downloaded at the SMRP website 6
C
Contents of the BoK
• The BoK contains the 5 pillars of the maintenance and reliability
community:
– Business and Management
– Manufacturing Process Reliability
– Equipment Reliability
– Organization and Leadership
– Work Management
• The CMRP test contains elements of each of these pillars
7
C
Manufacturing
Process Reliability
Webinar #2
Wednesday, July 16, 2014
8
M
Suggested Reading
http://reliabilityweb.com/ 9
C
Reliability is Built In Not Added On
• Losses of Production
– 40% are production related – equipment pressed past its limit
– 40% are design related
– 20% are maintenance related – equipment repaired incorrectly
• 70% of your maintenance costs are “set-in-stone” during the
design phase
10
C
Reliability Models
Pump 1 Pump 2 Reliability .882
Reliability .98 Reliability .90
• Reliability of assets in series
– multiply the reliability of each item in series to obtain an overall reliability
In this example .98 x .90 = .882
11
C
Reliability Models
Pump 1 Parallel Rule:
Reliability .90 + − ∗ =
.9 + .98 – (.9 x .98) = 1.88 - .882 = .998
Pump 2
Reliability .98
• Pump 2 is a spare to Pump 1.
• If Pump 1 or Pump 2 runs, sufficient capacity will be supplied to
the process
12
C
Reliability Models
Pump 1 Parallel Rule:
Reliability .90 + − ∗ =
.9 + .98 – (.9 x .98) = 1.88 - .882 = .998
Pump 2
Reliability .98 When P3 is added P3 = .9
, + − , ∗ =
Pump 3 .998 + .90 – (.998 x .9) = .9998
Reliability .90
• Pump 2 and 3 are spares to Pump 1.
• If Pump 1,2 or 3 runs, sufficient capacity will be supplied to13the
process
C
Reliability Models - HOMEWORK
Pump 1 Pump 2
Reliability .98 Reliability .97
Pump 3 Reliability?
Reliability .91
Pump 4
Reliability .88
• Calculate the reliability of this system and report answer
tomorrow. Pump 3 and 4 is a spare to pump 2.
• If pump 2,3, or 4 runs the system capacity will be satisfied.
14
Pump 1 must run for Pump 2 to run.
C
Three Separate Problems
• , = 1 ∗ 2
• .98 * .97 = .9506
• , + − , ∗ = ,,
• (.9506 + .91)-(.95 * .91) = 1.8606 - .8650 = .99554
• ,, + − ,, ∗ = ,, ,
• ((P1,2,3) + P4)-((P1,2,3) * P4)= P1,2,3,4
• (.99554 + .88)-(. 99554 * .88) = 1.87554 - .876088 = .999466 15
C
Plant Failure Reliability Logic Diagrams
COMP FAIL
COMP FAIL ASSET FAIL
COMP FAIL
COMP FAIL
COMP FAIL ASSET FAIL SYSTEM FAIL
COMP FAIL
COMP FAIL ... PROCESS FAIL
COMP FAIL ASSET FAIL
COMP FAIL
...
... PLANT FAIL
.
. 16
.
C
Overall equipment effectiveness (OEE)
17
C
OEE – Overview
• Productivity, Inc. began publishing books on Total Productive
Maintenance (TPM) in 1988.
• An early discussion of OEE was included in Seiichi Nakajima’s
book Introduction to TPM.
• OEE has become a standard tool for increasing equipment
output, equipment reliability, and product quality
• OEE is intended to be a de-bottlenecking tool
• It is often misunderstood and misused
• This presentation is intended to introduce you to OEE 18
C
Understanding OEE Calculation
• The Six Big Losses
Loss Categories The Six Big Losses
Downtime Equipment Failures
(Lost Availability) Setup and Adjustments
19
C
Understanding OEE Calculation
• The Six Big Losses
Loss Categories The Six Big Losses
Downtime Equipment Failures
(Lost Availability) Setup and Adjustments
Speed Losses Idling and Minor Stoppages
(Lost Performance) Reduced Speed Operation
20
C
Understanding OEE Calculation
• The Six Big Losses
Loss Categories The Six Big Losses
Downtime Equipment Failures
(Lost Availability) Setup and Adjustments
Speed Losses Idling and Minor Stoppages
(Lost Performance) Reduced Speed Operation
Defect Losses Scrap and Rework
(Lost Quality) Startup Losses
21
C
Understanding OEE
22
C
Understanding OEE
23
C
Availability Definitions
• Availability is the percentage of calendar time that a unit is
running (or available to run) regardless of run rate or yield
• Availability is Also Know As: There are differences of
– Utilization opinion
– Uptime
• Utilization and uptime (i.e. run time as a percentage of
scheduled operating time) are sometimes substituted for
availability in OEE calculations. They are rarely equivalent!
• Many companies may have their own definitions of availability.
24
Convert these figures to this standard definition before benchmark comparison.
C
Availability - Example
• Availability = Run Time/Hours Per Calendar Year x 100%
– A piece of equipment is operated 24 hours/day for 5 days/week.
– What is the availability of this equipment?
• Run Time =
– 24 hours/day x 5 days/ week x 52 weeks/year = 6,240 hours/year
• Hours in the Year = 8,760 hours/year
= ∗ % = , %
25
C
Availability Benchmarks
• Industry and internal benchmarks help to identify better/best
practices and configurations
• Best-of-the-Best benchmarks recognize differences in
continuous, batch, and discrete processes and promote
discovery of best-in-class processes
26
C
Benchmarks Example
100
90
80
70
60 Historical availability
50 Availability
of a power generating
40 station
30
20
10
0 Sep '01 Oct '01 Nov '01 Dec '01 Jan '02 Feb '02 Mar '02
Best-of-the-Best Benchmarks
27
C
Run Rate - Definitions
• Run rate is the percentage of the standard or potential
production rate achieved during a specific period of time
• Run rate is Also Know As: - Efficiency
– (check the formulas. Don’t assume.)
• Defective units produced are not eliminated when computing
run rate but downtime is eliminated
• Secure run rate data from production planning, production
management, or operating logs
28
C
Run Rate - Example
'()*+ ,-).-)
• " # "$%& = ∗ 100%
/()01)2*+ ,-).-) *) 3*)04 5.004
• A piece of equipment can produce 10,000 widgets per year.
The equipment produced 8,945 widgets in one year.
– What is the Run Rate?
789
• " # "$%& = ∗ 100% = 89,4%
::::
Note: Run Rate can be calculated using units per second, per minute, per hour,
per day, per shift, etc.
29
C
Run Rate – Why Important?
• Run rate improvements increase thru put and revenue without
additional fixed costs
• Decreasing run rate variability enables reductions in raw
material, work in process, and finished goods inventories
without increasing risk of missed customer promised dates or
downtime due to inadequate feed material
30
C
Yield - Definitions
• Process Industries
– Yield is the percentage of the mass or volume inputs that a production
unit produces as top quality output meeting all specifications.
• Discrete Manufacturing
– Yield is the percentage of produced units that meet all quality
specifications.
• Metals
– Prime yield is the percentage of first quality production compared to
theoretical optimums.
31
C
Yield - Also Know AS
• First pass yield
• First run quality rate
• Prime yield
• Recovery
• Pass rate
• Unavoidable scrap in the production process is sometimes
deducted from inputs and outputs for yield calculations
• Secure yield data from the quality control or production control
department
32
C
Yield - Why is it Important?
• Eliminating yield losses produces corresponding throughput
and revenue increases at minimal cost
• Eliminating yield losses reduces scrap, waste handling, and
environmental issues
• Eliminating variability in yield reduces production uncertainties
allowing better customer delivery with lower work in process
and finished goods inventories
33
C
Yield
B*CC (D B*)0E2*+C ,-).-)
• >?&@A = ∗ 100%
B*CC (D B*)0E2*+C F1.-)
'()*+ G24H0)C I00)21H C.0J2D2J*)2(1
• >?&@A = ∗ 100%
'()*+ G24H0)C .E(J0CC04 KL -12)
B*CC (D .E(4-J04 I*)0E2*+ I00)21H C.0JC
• >?&@A = ∗ 100%
B*CC (D I*)0E2*+ .E(4-J04 2D 1( E0M0J)C
• >?&@A = 100% − "&N&O% P$%&
• Where reject rate is percentage of input materials represented by
nonconforming outputs
34
C
Yield
'()*+ G24H0)C I00)21H C.0J2D2J*)2(1
• >?&@A = ∗ 100%
'()*+ G24H0)C .E(J0CC04 KL -12)
• A piece of equipment produced 785 widgets. There were 698 widgets
produced that meet the specifications.
– What is the Yield for this piece of equipment
Q87 G24H0)C
• >?&@A = ∗ 100%
R79 G24H0)C
• Yield = 88.9%
35
C
OEE – Equation
OEE is defined as:
STT = UV$?@$W?@?%X ∗ " # "$%& ∗ >?&@A
What is the OEE for the piece of equipment that was used as an example in
the last few slides?
STT = 71,2% UV$?@$W?@?%X ∗ 89,4% " # "$%& ∗ 88,9% (>?&@A)
STT = 56,5 %
Is an OEE of 56.5% good?
36
C
OEE – Benchmark
Generally Accepted Benchmarks for OEE are:
Continuous Process, e.g. bottling plant. 95%
Batch Process, e.g. polymer plant. 85%
37
C
OEE – Benchmark
(Courtesy of Benchmarking in the Process Industries, Ahmad & Benson)
38
C
OEE – General Comments
OEE is a statement and not a judgment
OEE is used to measure improvement
OEE calculations that have been “modified” to produce
artificially high OEE numbers are not suggested
Many control systems have OEE “built into” the software, or
software can be added to existing control systems to
automatically capture data and calculate OEE in real time
OEE should not be used to compare one piece of equipment or
process to another piece of equipment or process. 39
C
OEE – General Comments
Software can be purchased that will calculate and track OEE
OEE can be calculated using a standard spreadsheet e.g. XL
OEE was designed to be a de-bottlenecking tool
Develop codes to capture causes of decreased availability, run
rate or yield
Keep the causes to a minimum
Plot causes on Pareto charts and analyze often
Using this data, take action to improve availability, run rate, 40and
yield
C
Statistical Process Control - SPC
Statistical process control (SPC) is a method for achieving
quality control in manufacturing processes.
It is a set of methods using statistical tools such as mean,
variance and others, to detect whether the process observed is
under control.
Statistical process control was pioneered by Walter A. Shewant
and taken up by W. Edwards Deming with significant effect by
the Americans during World War II to improve industrial
production.
Deming was also instrumental in introducing SPC methods to
Japanese industry after that war. Dr. Shewhart created the
basis for the control chart and the concept of a state of
statistical control by carefully designed experiments 41
http://en.wikipedia.org/wiki/Statistical_process_control
C
Statistical Process Control - SPC
Classical Quality control was achieved by observing important
properties of the finished product and accept/reject the finished
product. As opposed to this technique, statistical process
control uses statistical tools to observe the performance of the
production line to predict significant deviations that may result
in rejected products.
By using statistical tools, the operator of the production line can
discover that a significant change has been made to the
production line, by wear and tear or other means, and correct
the problem - or even stop production - before producing
product outside specifications.
An example of such a statistical tool would be the Shewhart
control chart, and the operator in the aforementioned example
plotting the net weight in the Shewhart chart. 42
http://en.wikipedia.org/wiki/Statistical_process_control
43
C
Don't forget to put this in your agenda !
Webinar #3 - Wednesday, August 6, 2014
44
Charles Marks
Mob.: +1 (304) 591-2154
Email: Charles.Marks@Fluor.com
