Professional Documents
Culture Documents
TRAINING MANUAL
FOR
BY
Preface
Various studies have indicated that for large manufacturing systems or pieces of equipment,
maintenance and its support account for as much as 60 to 75 percent or more of their life
cycle costs. The increasing demands on high quality products have brought the maintenance
problem into even sharp focus. This, therefore, has put more emphasis on maintainability
during product design.
Terotechnology is the process of optimising the life cycle costs of an asset or equipment. In
the process of optimising life cycle costs a thorough understanding of plant reliability and
maintainability is very crucial. An attempt, therefore, has been made to present reliability
and maintainability concepts in this book to meet the challenges of modern manufacturing
system design. In doing so every effort has been made to treat the topics discussed in such a
manner that the reader will need minimum previous knowledge to understand the contents.
This introductory book describes how to design for plant reliability and ease maintenance
from the first stages of product design. In addition the book stresses how to:
• Improve product performance
• Minimize downtime
• Reduce the frequency of maintenance
• Lower the cost of maintenance
• Use life cycle costing in the process of choosing and purchasing equipment
• Decide when and how to replace an equipment
• Acquire data for reliability growth
The book is intended for Bachelor of Industrial Technology students and other engineering
students, as well as professional engineers and design and maintenance managers. This book
will enable manufacturing firms to stay ahead of the rest in product quality, efficiency and
profitability.
2
Contents
3
4.5.1 Availability and Scheduled maintenance................................................................39
4.5.2 Losses caused by non-availability of the system.....................................................39
4.6 Downtime and Maintenance Strategies..........................................................................40
4.6.1 Mean Downtime (MDT) and Mean time to repair (MTTR) [or Repair Time].......40
4.6.2 Active Repair Time.................................................................................................40
4.6.3 Factors Influencing Downtime................................................................................41
4.7 Comparisons of Maintainability and maintenance costs ...............................................42
4.8 Comparisons of Reliability and maintenance Costs.......................................................42
4.8.1 Factors affecting Reliability and Maintenance Costs..............................................43
4.9 Reliability – Centred Maintenance (RCM)....................................................................43
4.9.1 Basic steps in RCM Process....................................................................................44
4.9.2 Methods of Monitoring Equipment condition.........................................................46
4.9.3 The Benefits of RCM Application..........................................................................48
5. Life Cycle Cost and the Cost of an Equipment....................................................................49
5.1 The costs of quality........................................................................................................49
The Life Cycle Costs ...........................................................................................................49
..........................................................................................................................................52
5.2.1 Life Cycle Costing (LCC).......................................................................................52
5.2.2 Life cycle costing steps are:....................................................................................53
5.2.3 Advantages and Disadvantages of Life-cycle costing.............................................53
5.2.4 Why Use LCC?.......................................................................................................54
.............................................................................................................................................54
5.2.5 The Conversion or Decommission Phase of Life Cycle Costing ...........................54
5.2.6 Life Expectancies....................................................................................................55
5.2.7 Life Cycle Cost models...........................................................................................56
Example 1.................................................................................................................................58
Manufacturer B’s Electric generator....................................................................................59
5.3 Cost and Performance of Equipment.............................................................................60
5.3.1 Factors related to reliability.....................................................................................60
5.3.2 The Design Profile..................................................................................................61
6.0 Maintainability and Reliability terms and definitions........................................................61
7.0 Depreciation and Equipment Replacement........................................................................63
7.1 Economic Life and Obsolescence..................................................................................63
7.2 Depreciation...................................................................................................................64
7.3 Obsolescence..................................................................................................................64
7.4 Causes of Depreciation...................................................................................................64
Depreciation.....................................................................................................................64
7.4.1 Methods of calculating Depreciation......................................................................65
7.5 The Decision Whether to Purchase................................................................................67
7.6 Installation of new Equipment.......................................................................................68
7.7 Equipment Replacement................................................................................................68
7.7.1 Reasons for Replacement of Equipment.................................................................68
7.7.2 Equipment Replacement Policy..............................................................................69
7.7.3 Guidelines in Replacement Analysis.......................................................................69
7.7.4 Methods used for Replacement...............................................................................69
Solution................................................................................................................................70
Machine A................................................................................................................................71
Machine B................................................................................................................................71
(f) MAPI Method.............................................................................................................72
8.0 Acquisition of Failure Data ...............................................................................................72
8.1 Reasons for data collection..........................................................................................73
8.2 Information and Difficulties...........................................................................................73
8.3 Best Practice and Recommendations.............................................................................75
4
8.4 Analysis and Presentation of Results.............................................................................76
8.5 Sources of Reliability Information.................................................................................76
1. Terotechnology
1.1 Introduction
Terotechnogy is the process of optimising the life cycle cost of physical assets. Life cycle
cost is the sum of all costs incurred during the life time of an asset that is, the total of
procurement and ownership costs. Life cycle costs are categorised as: cost of acquisition, cost
of use, and cost of administration. Life Cycle Cost (LCC) of any physical asset is influenced
by the plant reliability and plant maintainability.
Maintainability is the action taken during the design and development of assets to include
features that will increase ease of maintenance and will ensure that when used in the field the
asset will have minimum downtime and Life-cycle support costs i.e. its serviceability,
reparability, and cost-effectiveness of maintenance are increased.
Reliability is the probability that an item will carry out its stated function adequately for the
specified time interval when operated according to the designed conditions, i.e. to define
reliability of any equipment:
• We must state the planned working life e.g. a new car might be very reliable if we
only expect it to last for 5 years; less reliability over a period of 10 years; and
completely unreliable if we are expecting a useful life of say 40 years.
• Similarly we shall need to know the intended conditions of use, and the routine
maintenance which is required, e.g. if a car engine seizes because there is no water in
the radiator this is a failure of maintenance rather than a failure of reliability; if a car
is driven carelessly and fails this is a misuse failure.
Mean Time between Failures (MTBF) and Mean Time to Repair (MTTR): The Mean
Time between Failures (MTBF) tells us how long on average, equipment operates before it
fails, and this we want to be as long as possible. MTBF, therefore, depends on reliability.
The Mean Time To Repair (MTTR) tells us how long on average, it takes to put the
equipment right after it has failed, and this we want to be as short as possible. MTTR,
therefore, depends on maintainability.
2. Plant Reliability
Reliability is the probability that an item will carry out its stated function adequately for the
specified time interval when operated according to the designed conditions. Since no two
equipments are identical due to manufacturing differences however the designer and control
engineers try to eliminate any defects, reliability is given in percentages (for mathematical
reasons, it is expressed in decimals of 1.00). Suppose that out of every 100 cars of a
particular type, 99 prove to be trouble free if used and maintained correctly, and one fails to
work as intended. Then we can say that the reliability of each car is 99 percent, meaning that
the chances are 99 in 100 that it will prove reliable.
The longer we expect anything to last the more likely it is to fail during that time i.e.
reliability falls as time increases.
5
Reliability at time t = R(t) = (No. surviving at instant t)/ (No. at start when t=o)
The Reliability of large installations is not necessarily quoted as a single figure. The central
unit, for example, may be assigned a higher reliability than some of the ancillary and support
equipment. Ideally, the minimum acceptable reliability or MTBF or the maximum failure rate
should be quoted, but when the system is of very new design it may be more realistic to get a
target reliability, MTBF or failure rate. In deciding what this should be, we must consider the
conditions of use, the duty cycle, what the maintenance requirements are likely to be and how
long the system can be out of use while maintenance is done.
Example: Consider a transformer and rectifier set, used to convert mains electricity to a
suitable voltage and frequency. Suppose each part has a reliability of 0.9. If we require only a
transformer and nothing else, then the system will have the same reliability as the one part it
contains. Therefore, for 1 part Reliability = 0.9
If however we require a rectifier, then we have two things, which can go wrong. Therefore,
for 2 parts Reliability = 0.9 x 0.9 = 0.81
and for 3 parts reliability = (0.9)3 = 0.73
and for 10 parts reliability = (0.9)10 = 0.35
6
Reliability is the probability that an item will perform as required, under stated conditions, for
a stated period of time. Hence since performance is an aspect of quality, we might say that
reliability is the probability an item will retain its quality, under stated conditions, for stated
period of time. Thus quality and reliability are very closely related. Hence the quality of a
product from the manufacturer will affect its reliability. The quality of the product is also
affected by:
(i) the method of manufacture
(ii) Production equipment
(iii) Inspection and test equipment
(iv) Supplies and/or selection of raw materials and parts etc.
(All these assume that the design and development of the product has been done
correctly).
Two major areas of engineering activity determine the reliability of an engineering system.
First, provision for reliability must be established during the earliest design concept stage,
carried through the detailed design development, and maintained during the many steps in
manufacture. Once the system becomes operational, it is imperative that provision be made
for its continued maintenance during its service.
7
i) Margin of safety
Variability in the strength properties of materials and in loading conditions (stress) leads to a
situation in which the overlapping statistical distributions can result in failures. Therefore,
variability in strength has a major impact on the probability of failure, so that failure can be
reduced with no change in the mean value if the variability of the strength can be reduced.
ii) Derating
The analogy to using a factor of safety in structural design is derating electrical, electronic,
and mechanical equipment. The reliability of such equipment is increased if the maximum
operating conditions (power, temperature, etc.) are derated below their nameplate values. As
the load factor of equipment is reduced, so is the failure rate. Conversely, when equipment is
operated in excess of rated conditions, failure will ensue rapidly.
iii) Redundancy
Redundancy is the most effective way of increasing reliability. In parallel redundant designs
the same system functions are performed at the same time by two or more components even
though the combined outputs are not required. The existence of parallel paths may result in
load sharing so that each component is derated and has its life increased by a longer than
normal time.
Another method of increasing redundancy is to have inoperative or idling standby units that
cut in and take over when an operating unit fails. The standby unit wears out much more
slowly than the operating unit does. Therefore, the operating strategy often is to alternate
units between full-load and standby service. The standby unit must be provided with sensors
to detect the failure and switching gear to place it in service. The sensors and/or switching
units frequently are the weak link in a standby redundant system.
iv) Durability
The material selection and design details should be performed with the objective of
producing a system that is resistant to degradation from such factors as corrosion, erosion,
foreign object damage, fatigue, and wear. This usually requires the decision to spend more
money on high-performance materials so as to increase service life and reduce maintenance
costs. Life cycle costing is the techniques used to justify this type of decision.
v) Damage tolerance
Crack detection and propagation have taken on great importance since the development of the
fracture mechanics approach to design. A damage-tolerant material or structure is one in
which a crack, when it occurs, will be detected soon enough after its occurrence so that the
probability of encountering loads in excess of the residual strength is very remote.
vii) Simplicity
Simplification of components and assemblies reduces the chance for error and increases the
reliability. The components that can be adjusted by operation or maintenance personnel
8
should be restricted to the absolute minimum. The simpler the equipment needed to meet the
performance requirements the better the design.
viii) Specificity
The greater the degree of specificity the greater the inherent reliability of design. Whenever
possible, be specific with regard to material characteristics, sources of supply, tolerances and
characteristics of the manufacturing process, tests required for qualification of materials and
components, procedures for installation, maintenance, and use. Specifying standard items
increase reliability. It usually means that the materials and components have a history of use
so that their reliability is known. Also, replacement items will be readily available. When it
is necessary to use a component with a high failure rate, the design should especially provide
for the easy replacement of that component.
The malfunctions that an engineering system can experience can be classified into five
general categories.
1. Design mistakes: Among others the common design errors are failure to include
all important operation factors, incomplete information on loads and
environmental conditions, erroneous calculations, and poor selection of materials.
2. Manufacturing defects: Although the design may be free from error, defects
introduced at some stage in manufacturing may degrade it. Some common
examples are (1) poor surface finish or sharp edges (burrs) that lead to fatigue
cracks and (2) decarburization or quench cracks in heat-treated steel. Elimination
of defects in manufacturing is a key responsibility of the manufacturing
engineering staff, but a strong relationship with the R&D function may be
required to achieve it. Manufacturing errors produced by the production work
force are due to such factors as lack of proper instructions or specifications,
insufficient supervision, poor working environment, unrealistic production quota,
inadequate training, and poor motivation.
3. Maintenance: Most engineering systems are designed on the assumption they
will receive adequate maintenance at specified periods. When maintenance is
neglected or is improperly performed, service life will suffer. Since many
consumer products do not receive proper maintenance by their owners, a good
design strategy is to make the products maintenance-free.
4. Exceeding design limits: If the operator exceeds the limits of temperature, speed,
etc., for which it was designed, the equipment is likely to fail.
5. Environmental factors: Subjecting equipment to environmental conditions for
which it was not designed, e.g., rain, high humidity, and ice, usually greatly
shortens its service life.
9
reliability. The summation of these two curves produces the total cost curve, which has a
minimum at an optimum level of reliability. Other types of analyses establish the optimum
schedule for part replacement to minimize cost.
Total Cost
Cost
Cost of Design
And Manufacture
Costs after Delivery
Rm
Reliability
Figure: Influence of Reliability on Cost
Rm – Optimum Reliability
(ii) Prepare the design and express it as a manufacturing specification – The designer
must specify what has been made in order to satisfy the requirement of the customers.
(iii) Prove the design – Wherever the design is a departure from previous practice, we
shall need tests on prototypes to show by demonstration that what is proposed will
achieve the reliability demanded.
(vi) Purchase, storage, installation and commissioning of new equipment – we next look at
reliability form the customer’s point of view, and consider how he decides what to
purchase and how it is stored, installed and commissioned for use.
(vii) Operation and maintenance – The customer must use and maintain the equipment as
intended, employing operators with adequate skills and training. If there are any
difficulties, the manufacturer should be anxious to help, not merely in the interests of
good customer relations but also in order to learn for the future.
1
(viii) Reliability management and Prediction – Finally we consider the overall management
of reliability, and how reliability of a proposed design can be predicted.
(a) Definition:
When an item no longer works as intended we say it has failed. Failure, therefore, is the
termination of the ability of an item to perform its required function.
i) Cause –
• A misuse of failure is a failure attributable to the application of stresses
beyond the stated capability of the item i.e. ill treated.
• An Inherent Weakness failure is a failure attributable to weakness inherent
in the item itself, when subjected to stresses within the stated capabilities of
the item i.e. failure is probably due to a design or manufacturing fault.
ii) Suddenness –
• A sudden failure is one which could not be anticipated by prior examination.
• A gradual failure is one, which could be anticipated by prior examination i.e.
it is possible to predict that it will occur since it takes place gradually.
iii) Degree –
• A partial failure is one resulting from deviations in characteristics beyond
specified limits, but not such as to cause complete lack of the required
function i.e. the item does not work as well as it should, but it has not
completely failed.
• A complete failure is one resulting from deviations in characteristics beyond
specified limits, such as to cause complete lack of the required function.
iv) Combination of the above terms –
• A catastrophic failure is one which is both sudden and complete.
• A degradation failure is one, which is both gradual and partial.
(c) Failure rate
Failure rate is the probability of failure in unit time of an item, which is still working
satisfactorily.
From a sample of 1000 parts, suppose 100 hours after the start of a test we notice that 221
parts have failed, leaving 779 still working. Then:
Observed Reliability over 100 hours = (No. Surviving at t= 100)/(No. at start of the test at
t=0) = 779/1000 = 0.779; After the test had run 200 hours the number of failures might have
risen to 400, and then: Observed Reliability over 200 hours = 600/1000 = 0.600
Suppose that at the instant when t= 200 hours, we are able to find out that parts are failing at
exactly 1.5 per hour. Since 600 are still working, we can express the rate at which they are
failing as: Failure rate 8 = (No. failing per hour at instant t)/ (No. still surviving at instant t)
8 = 1.5/600 = 0.0025 per hour =25 x 10-4 per hour.
1
above, the parts were failing at exactly 1.5 per hour at the instant when t = 200 hours, so we
can write: Observed value of failure probability density function for instant t =
(No. failing per hour at instant t)/(No. at start (t=0)
:. F(t) = 1.5/100 = 15x 10-4 per hour, and total f(t) = 1.00.
If after the test had run 200 hours the number of failures might have risen to 400, and then we
should have; Observed Reliability over 200 hours = 600/1000 = 0.6.
Suppose that at the instant when t=200 hours, we are able to find out that parts are failing at a
exactly 1.5 per hour; since 600 are still working, we can express the rate at which they are
failing as follows: Observed instantaneous failure rate = (Number failure per hour at instant
t)/(Number still surviving at instant t)
Therefore, Instantaneous failure rate (or Hazard Rate) λ =1.5/600=25x10-4 per hour.
Having counted how many parts failed during each hour of the test, we could have said that
x1 per cent of the 1000 parts failed during the first hour, x2 per cent during the second hour,
and so on up to say, xn per cent during the hour when the 1000th part failed. Since all the parts
have now failed, if we add x1+x2+x3+……..+xn, the total must be 100% or 1.00.
At the instant t=200 hour, parts were failing at exactly 1.5 per hour:
Observed value of failure probability density function for instant t =
(Number failing per hour at instant t)/(Number at start when t=0).
Therefore, f(t)=1.5/1000=15x10-4 per hour.
If we add up the fraction, which fail for every instant of time, the total must be 1.00. For a
continuous mathematical function we can write this as
Reliability is cumulative. When we say that reliability of 100 parts over the period time t=0,
to time t=t1, was observed to be 0.70, we mean that if we add up all those which failed
between the start of the test and time t, there would be a total of 30 failures out of 100.
Suppose we continued to a new time t2, and during this extension a further 3 parts failed, so
that the reliability is now 0.67, therefore:
1
= 0.67 - 0.70 = -0.03 (The negative sign means that the reliability is reduced). But this is the
observed value of failure probability density function for instant t, i.e. (Number failing per
hour at instant t)/(Number at start when t=0) = 3/100 = 0.03.
Therefore, the value of the failure probability density function f(t) = -(the rate of change of
reliability), i.e. f(t)= -dR(t)/dt=λR(t); Therefore, dR(t)/R(t)= -λdt, or ƒdR(t)/dt= t= 0ƒt –λdt,
Therefore, lnR(t)=-λt; or R(t)=exp(-λt), and f(t)= λR(t)=λexp(-λt)
Example: A part has a constant failure rate of 0.001 per hour. Calculate its reliability over
500 operating hours.
Solution: R(t)=exp(-λt)=exp(-0.001x500)=exp(-0.5)=0.61.
We can also calculate the fraction of parts originally put on test, which are failing per hour at
this instant when the time is 500 hours, since
f(t)= λR(t)=λexp(-λt)=0.001x0.61=0.00061=61x10-4 failures per hour.
(f) Relationship between Failure Rate λ, and Failure Probability Density Function
f(t)
Consider the probability of an item failing in the interval between t and t+dt. This can be
described in two ways:
(a) The probability of failure in the interval t to t+dt given that it has survived until time
t which is λ(t)dt; where λ(t) is the failure rate.
(b) The probability of failure in the interval t to t+dt unconditionally, which is f(t)dt,
where f(t) is the failure probability density function.
The probability of survival to time t is the reliability R (t). The rule of conditional probability
therefore dictates that: λ(t)dt=f(t)dt/R(t); therefore, λ(t)=f(t)/R(t).
However, if f(t) is the probability of failure in dt then:
t= 0 ƒtf(t)dt=probability of failure 0 to t=1-R(t)
t
t= 0ƒ f(t)dt =1-R(t), or f(t)= -dR(t)/dt, therefore, λ(t)= -(dR(t)/dt)/R(t)
t R(t)
t= 0ƒ λ(t)dt= - ƒ
1 dR(t)/R(t)= -lnR(t);
[NB when t=0, R(t)=1 and at t the reliability is R(t)]
[NB: Failure Rate, λ, is the probability of failure in unit time of an item, which is still
working satisfactorily, i.e. 1.5/600=25x10-4 failures per unit time. Whilst, Failure Probability
Density Function, f(t), is the probability of failure in unit time of an item which was working
satisfactorily at time t=0, i.e. 1.5/1000=15x10-4 failures per unit time].
1
period
8 = Failure rate
O A 8 B C
At the start of the test the failure rate may be relatively high, but this usually falls
progressively, until at A where the failure rate is approximately constant and at its lowest
level. The most common causes of early failures are:
• Manufacturing faults – these are faults which are not detected before dispatch to the
consumer. In each case two faults are implied: the product was wrongly made; and its
fault was not detected before it left the factory. Manufacturing faults often account for
the majority of the early failures.
• Design faults – when the designer completes a new design, it must be thoroughly
tested as a proto type before full-scale manufacture begins. If this is not adequately
done, any shortcomings in the design may then reveal themselves as early failures.
• Misuse – A few failures may be due to accidental customer misuse, before he/she is
fully competent in operating the product.
Increasing the early failure period prior to dispatch, making improvements in the
manufacturing process, and improving quality control activities can all minimize the
occurrence of early failures. Some of the reasons for failures in this period include
substandard workmanship and parts, poor manufacturing methods, human error, inadequate
quality control, and unsatisfactory debugging.
NB: This is the period often covered by guarantee, during which the manufacturer agrees
to make good anything, which goes wrong.
ii) Constant failure period: A-B
Once the early failures have been removed, the parts usually settle down to what may be a
relatively long period, when the failure rate is approximately constant, from A to B, after
which the failure rate begins to rise again, often quite steeply as the parts begin to wear out.
Although the failure rate in the constant period is usually low, it can be very troublesome if
high reliability is required. We can avoid early failures by good design and manufacture, and
by running the parts on load for a time at least equal to OA. We can avoid wear out failures
by replacement before time B, but we are still left with the constant failure rate period, right
through the normal working life. Failures in this period are unlikely to be due to any single
cause, it is usual for failures from a wide variety of causes to occur at random, with no
obvious pattern, except that the failure rate is roughly constant.
Some of the reasons for failures in this period are undetectable defects, low safety factors,
high-unexpected random stress, abuse, and natural failures.
1
(h) System
A system is used to denote any complete installation or equipment. The failure pattern of a
system, or indeed any assembly of parts, can be regarded as the sum of the failure patterns of
the individual parts, but there may be additional failures as follows:
(i) Failures may occur at what are termed the interfaces between two parts. For
example there may be failures in soldered jointed or connectors, as well as in the
parts themselves.
(ii) One part may affect the performance of another. Thus if one part fails, it may
overload other perfectly good parts and cause them to fail as well.
Whenever a system fails we repair it, probably by replacing the faulty part with a new one.
When repair is no longer economical, we buy a new system. Thus each system will be a
different age, and each of the large number of parts it contains will have its own failure
pattern. The addition of so many small failure patterns is likely to produce a roughly constant
failure rate for the whole system.
2.10 Redundancy
1
• in hospitals there is always an emergency supply (in case the mains electricity fails),
often from a standby motor generator;
• a bicycle wheel has several spokes, several can break without a serious drop in
performance or reliability. This is called partial redundancy;
• a spare wheel is provided for vehicles in case one is punctured etc.
The electrical analogy is that current will flow from A to B so long as there is an unbroken
circuit through at least one unit. The probability that at least one of them is still working will
give the overall reliability of the units (in parallel).
Thus although one unit alone would only be 0.90 or 90% reliable, three units in parallel are
0.999 or 99.9% reliable. For reasonable gain in reliability there is a limit to how units you
can put in active redundancy. Thus:
1
• Two units in parallel will in many cases give a useful improvement in reliability. If
however we insert a third unit, the additional improvement is much smaller, and in
general there is no advantage in putting large numbers of units in parallel.
• We get the biggest increase when the unit reliability is around 0.35 and 0.50. If the
unit reliability is already 0.85 or above, two units in parallel may be useful, but the
use of three upwards gives little further improvement.
• However, when the unit reliability is very low, it is difficult to get acceptable block
reliability from the use of redundancy alone because if our units are unreliable the use
of more of them only puts in more unreliability and this offsets the gain from
redundancy.
R4=
.90
R1=
.95
R5=
.95
R2= R3=.80
.80
R6= R8=
.85 .85
R7= R9=
.85 .85
1
• Look for units in series within a parallel configuration, and calculate the reliability of
a single equivalent unit.
• Reduce each parallel arrangement to a single equivalent unit.
• Repeat the above, always dealing with the smallest recognisable configurations first.
R1= R4=
.95 .90
R5=
.95
R23=
.96 R6789=
.92
B
• Next take each series limb in turn; R123 = R1 x R23 = ).95 x 0.96 = 0.91
R6789 = 0.90 x 0.95 x 0.92 = 0.79
We now have a straightforward parallel configuration; F123 = (1.0 – 0.91) = 0.09
F456789 = (1.0 – 0.79) = 0.21
Therefore, the probability the whole system will fail = Fs = 0.09 x 0.21 = 0.0189, and
the reliability of the whole system = Rs = (1.0 – 0.0189) = 0.9811
A A
RB1 RB2
RB1 RB2
B
B
1st alternative: 2nd alternative:
A parallel arrangement A parallel arrangement
Without connections With connections
1
Suppose every unit has a reliability of 0.90:
With 1st alternative - the whole system will work if either:
RA1 and RB1 both work, or RA2 and RB2 both work. However there are two more possible
combinations, which are apparently unacceptable, namely:
We could make both these latter combinations acceptable merely by modifying the
configuration to the 2nd alternative in which there is a cross connection in the middle of the
network.
Reliability of the whole system in the 1st alternative = 0.964 and reliability of the whole
system in the 2nd alternative = 0.9801.
Thus the 2nd alternative is better than the 1st alternative for the same units, in the same
application.
(i) The standby unit is not carefully packed away in a specially designed storeroom.
Usually it is out on the job, alongside the unit, which is operating. Therefore it can
be affected by the environment due to, say vibration, grease, dust etc. Thus the
assumption that it will remain in perfect working order, ready for immediate use,
may be quite incorrect. It may well have an appreciable failure rate while waiting
to be used, and the longer it remains on standby duty the greater the risk that it
will not work should it be required.
(ii) For many systems, the moment of switch on is particularly hazardous. Stray
surges are liable to wander around the circuits and may well damage any part,
which is already a bit doubtful. With active redundancy, however, the second
circuit is switched on with the first and there is time to check that it works before
it is required.
(iii) Some parts deteriorate if left unloaded for long periods. Large electric furnaces
are best kept at least partially energized, since heater failures often occur if for any
reason they have to be completely switched off and allowed to cool.
(iv) It may be possible for two units in active redundancy to share the work, so that
each only has to work at half load. This effectively de-rates both of them, and so
may lead to a longer working life.
1
(v) In many cases the operator may not notice the failure and starts the standby unit.
Usually it will be necessary to install further devices to:
(a) detect that the main unit has failed
(b) Switch in the standby automatically.
If either a or b above do not work correctly, the standby unit will not come into
operation, even though it is in perfect working order.
The detection device will normally be energized whenever the main unit is working, and
there are two obvious ways in which it can fail:
(a) it can fail to detect that the main unit has ceased to function
(b) It can cause the standby unit to switch in when the main unit is still working
correctly.
The switchover device will not normally be energized, but it must function correctly when
required. It can also fail in two ways:
(a) it can fail to switch in the standby unit when required
(b) It can switch in the standby unit when it is unnecessary.
Clearly in choosing between active and standby redundancy, we shall have to consider each
individual case in its merits, but much often hinges on two factors:
(i) The reliability of the detection and switching device, of the standby unit.
We use redundancy because we are not satisfied with the reliability of one
working unit on its own. Unless therefore the detection and switching device
are appreciably more reliable than the main working unit, we may be
introducing as much unreliability into the system as we are removing by
having the standby unit.
(ii) It is a fundamental assumption in using a standby unit that its failure rate when
shut down, but probably exposed to normal operating conditions, is
appreciably lower than if it were energized under the same conditions. If this
is not so, then active redundancy may well be on better proposition.
2
3.0 Maintainability
Maintainability is the action taken during the design and development, and installation of a
manufactured product to include features that will increase ease of maintenance, reduce
required man-hours, tools, logistic costs, skill levels and facilities and ensure that when used
in the field the product will have minimum downtime, and life-cycle support costs.
From this definition, the general principles maintainability, therefore, include lowering or
eliminating altogether the need for maintenance, reducing life cycle maintenance costs,
lowering the number, frequency, and complexity of required maintenance tasks; establishing
the extent of preventive maintenance to be performed; reducing the mean time to repair
(MTTR); and providing for maximum interchangeability.
On the other hand maintenance refers to the measures taken by the users of a product to keep
it in operable condition or repair it to operable condition.
When maintainability engineering principles have been applied effectively to any product, the
following results can be expected.
• Reduced downtime for the product and consequently an increase in its operational
readiness or availability.
• Efficient restoration of the product’s operation condition when random failures are
the cause of downtime.
• Maximizing operation readiness by eliminating those failures that are caused by
age or wear-out.
Because equipment downtime consists of many components and sub-components, there are
numerous engineering and analytical efforts required to reduce downtime. The three main
components of equipment downtime are logistic time, administrative time, and active repair
time.
1. The three main components of equipment downtime are logistic time, administrative
time, and active repair time.
(a) Logistic time is that portion of equipment downtime during which repair work
is delayed because a replacement part of other component of the equipment is
not immediately available. Logistic time, therefore, is largely a matter of
management. By developing effective procurement policies can minimize it.
(b) Active repair time is that portion of equipment downtime during which the
repair staff is actively working to effect a repair. Its six elements are fault
location time, preparation time, failure verification time, actual repair time,
2
part acquisition time, and final test time. Usually, the length of active repair
time reflects factors such as product complexity, diagnostic adequacy, nature
of product design and installation, and the skill and training of the
maintenance staff.
(c) Administrative time is that portion of equipment downtime not taken into
consideration in action repair time and in logistic time. This time (that
normally include wasted time) is a function of the structure of the operational
organization and is influenced by factors such as work schedules and the non-
technical duties of maintenance people.
In many cases, the cost of acquiring a product is less than the cost of ownership over the
product life cycle. Cost of ownership includes operation costs (such as the cost of personnel,
facilities, and utilities), maintenance costs, the cost of test and support equipment, retirement
and disposal costs, technical data costs, the cost of training operations and maintenance
personnel and the cost of spares, inventory, and other support materials.
2
A cost effective and supportable design must take into account the maintainability
considerations that arise at each phase in the life cycle of the system or product. Careful
planning and systematic effort are needed to bring attention to important maintainability
design factors such as maintainability allocation, maintainability evaluation, maintainability
and design characteristics; maintainability parameters, and maintainability demonstration.
Each of these factors involves various sub-factors – e.g. packaging, standardization, inter-
changeability, human factors, safety, and testing and check out all play a role in the final
products maintainability design characteristics in every aspect of maintainability design.
The maintainability design characteristics are the features and design characteristics that help
reduce downtime and enhance availability. The goals of maintainability design include
minimizing preventive and corrective maintenance tasks; increasing ease of maintenance.
Decreasing support costs; and reducing the logistical burden by decreasing the resources
required for maintenance and support, such as spare parts, repair staff, and support
equipment.
The most frequently addressed maintainability design factors, ranked in descending order,
are: accessibility, test points; controls; labelling and coding; displays; manuals; check lists,
chart and aids; test equipment; tools; connectors; cases; covers and doors; mounting and
fasteners; handles; and safety factors; Other factors are standardization, modular design,
inter-changeability ease or removal and replacement, indication and location of failures,
illumination, lubrication, test adapters and test hook ups, servicing equipment, adjustments
and celebrations installation, functional packaging, fuses and circuit breakers; cabling and
wiring, weight, training requirements, skill requirements, required number of personnel, and
work environments.
(a) Standardization
This important design feature restricts to a minimum the variety of parts and components that
a product system will need. Standardization should be a central goal of design, because the
use of non-standard parts may lead to lower reliability and increased maintenance.
Some of the primary goals of standardization include: maximizing the use of common parts
in different products, minimizing then number of different types of parts, components,
assemblies and other items; maximizing the use of inter-chargeable and standard or off-the-
self-parts and components; minimizing the number of different models and makes of
equipment in use; controlling and simplifying inventory and maintenance; reducing storage
problems, and the effort spent on part coding and numbering.
(b) Inter-changeability
This is an important maintainability design factor that is made possible through
standardization. Inter-changeability means that, as an international aspect of design, any
component, part, or unit can be replaced within a given product or piece of equipment, by any
similar component, part, or unit. There are two types of inter-changeability; functional inter-
2
changeability and physical inter-changeability. In functional inter-changeability, two
specified items serve the same function. In physical inter-changeability, two items can be
mounted, connected and used effectively in the same locations and the same manner.
(c) Modularisation
Modularisation is the division of a system or product into physically and functionally distinct
units to allow removal and replacement. Each system or sub-system, from the highest to the
lowest level, can be designed as a removable entity. Questions of cost, practicality, and
function dictate the degree of modularisation. However, modular construction will reduce
training costs or provide other concrete benefit. Modularisation allows use of disposable
modules, which are designed to be discarded other than required once they fail (because
repair is either costly or impractical).
(d) Simplification
Probably the most difficult element of maintainability to achieve, but the most important, is
simplification. Simplification should be the constant goal of design. Even a complex product
or piece of equipment should appear simple and straightforward to the user. A good designer
incorporates important functions of a product into the design itself and uses as few
components as sound design practices will allow.
(e) Accessibility
Accessibility is the relative ease with which a part or piece of equipment can be reached for
service, replacement, or repair. Lack of accessibility is an important maintainability problem
and a frequent cause of ineffective maintenance.
(f) Identification
Adequate labelling or marketing of parts, controls, and test points facilitates maintenance
tasks such as replacement and repair. If a repair person is unable to readily part points, or
controls, maintenance tasks become more difficult, take longer to perform, and are more
likely to be performed incorrectly.
There are three distinct types of FMECA: System level, design level, and process level
FMECA. Of these three levels, the highest level of analysis is the system level FMECA,
which usually consists of a collection of subsystem FMECAs.
Performed in the initial design concept phase, the system level FMECA highlights potential
system or subsystem failures so that they can be prevented. The design level FMECA helps
identify and prevent failures stemming from the product design. It analyzes the design that
has been developed and examines how failures of individual items would affect the system
functioning or operation. The purpose of the process level FMECA is to analyze the process
by which the product or system is to be built and assess how potential failures in the
2
manufacturing or service process would affect the product/system functioning or operation.
All the three types of FMECA consist of the following basic steps:
• Understanding system parts, operation, and mission.
• Identifying the hierarchical, or indenture, level at which the analysis is to be
performed.
• Defining each item expected to be analysed – for example, component, module, or
subsystem.
• Establishing associated ground rules and assumptions – for example, system mission
and operational phases.
• Identifying possible failure modes for each item.
• Determining the effect of each item’s failure for every possible failure mode.
• Determining the effect of group failures – failures of more than one item – on system
operation and mission.
• Identifying methods, procedures, or approaches for detecting potential failures.
• Determining any provisions or design changes that would prevent failures or mitigate
their effects.
In using this analysis, the effect identified can be quite different depending on the objective
of the analysis. For example:
• In Reliability Analysis. The effect considered is the effect on the system’s or
equipment’s performance or ability to function.
• In Maintainability Analysis. The effects considered include the symptoms through
which failure to be pinpointed and the components that will require replacement as
the result of the failure.
2
• In Safety Analysis. The effects to be considered are damage to other systems and
equipment and possible danger to people.
Some of the advantages of the FMEA method are that it employs a systematic procedure to
categorize hardware failure and identifies all possible failure modes and their effects on
performance, personnel, and equipment. It is useful for comparing design, simple to
understand, and helps identify methods of detecting the various possible failures.
During the equipment design phase, professionals have many methods at their disposal for
eradicating or minimizing hazards to people and equipment. The basic objective of all these
approaches is hazard identification and control. Some of the safety analysis techniques are
hazard analysis, failure mode and effects analysis, and fault tree analysis.
Hazard Analysis Method – This safety analysis tool determines the safety requirements for
people, procedures, and equipment used in testing, operations, maintenance, and logistic
support. This method also determines the compliance of system and equipment with specified
safety requirements and criteria. (For FMEA and FTA see elsewhere)
2
• Providing correct tools and making regular adjustments to safety equipment
• Developing effective support procedures
• Proper inspecting of all tasks
• Making each individual conscious of hazards involved in his/her assignments
• Making workers safety conscious
• Proper training for performing tasks.
2
• If the equipment generates excessive heat, it is in principle better to design so that
it is not dissipated in the direction of the operator, rather than to leave the
customer to provide protective clothing. The latter is usually uncomfortable, and
so may not always be worn.
(d) Indicators:
Include dials, gauges, lamps bells, etc. which will show the working the working state of the
equipment. The majority are either visual or aural. Thus a light may come on or a bell may
ring to warn the operator that a particular cycle of manufacture is complete. A light to show
a particular part of system is on or off is cheaper and easier to observe. However, a meter
conveys more information and therefore it must be specified whenever an operator needs it.
Ensure that meters are calibrated directly in the units to be observed. A dial should be
calibrated so that it can be easily read to the required accuracy. Digital type maters, which
display actual numbers, are preferable as they are quicker to read and less prone to operator
error.
2
NB: Must controls tend to use a mixture of mechanical and electrical/electronic
principles.
2
ii) Fixing devices must be easy to release ad reassemble.
iii) Thus nuts, bolts and screws in awkward places should be avoided. It is difficult to
insert or tighten a screw into a position, which on e cannot see or perhaps cannot even
feel directly with the fingers.
iv) Plugs and sockets are attractive, because they make dismantling and reassembly
easier, although their own reliability may present a problem
v) Make sure that visual inspection is as ease as possible, and that parts such as nuts and
screws do not fall down inside when released immediately fall down inside.
vi) Where appropriate, make it possible to replace a complete unit, so that the faulty
one can be taken away for repair.
Thus the objective should be to make manuals as much as possible to be unnecessary or with
very little text by:
• using clear diagrams wherever they would be useful. A simple sketch often saves a
lot of writing, which the operator may not fully understand.
• Trying to put a diagram on the same page as the text to which it relates
• Keeping the wording concise and simple. Avoid technical terms which may mean
little to the operating and maintenance staff.
• Making sure that it is easy to locate any piece of information which may be required.
• Provide a good cross-referenced index, so that the users can find what they want even
though their terminology may be somewhat different from that used by the
manufacturer.
3
Spares parts and replacement parts should be stocked and supplied for models in current
production as well as for those which have gone out of production but are still used by
customers.
Serviceability is concerned with the ease with which maintenance can be performed on a
product. Many products require some form of maintenance or service to keep them
functioning properly. Products often have parts that are subject to wear and that are expected
to be replaced at periodic intervals.
It is important to anticipate the required service operations during the design of the product.
Provision must be made for disassembly and assembly. For example a design that requires
the removal of a body panel of an automobile to access the oil filter is inappropriate. It should
also be remembered that service usually will be carried out in "the field" where special tools
and fixtures used in factory assembly will not be available.
A concept closely related to serviceability is testability. This is concerned with the ease with
which faults can be isolated in defective components and subassemblies. In complicated
electronic and electromechanical products, testability must be designed into the product.
The best way to improve serviceability is to reduce the need for service by improving
reliability. Reliability is the probability that a system or component will perform without
failure for a specified period of time. Failing this, the product must be designed so that
components that are prone to wear or failure, or which require periodic maintenance, are
easily visible and accessible. It means making covers, panels, and housings easy to remove
and replace. It means locating components that must be serviced in accessible locations. If
possible press fits, adhesive bonding, riveting, welding, or soldering for parts that must be
removed for service should be avoided. Modular design is a great boon to serviceability.
Buildings and grounds is responsible for the appearance and functioning of buildings, parking
lots, lawns, fences, and the like. Equipment maintenance is responsible for maintaining
machinery and equipment in good working condition and making all necessary repairs. The
goal of maintenance is to keep the production system in good working order at minimal cost.
Decision makers have two basic options with respect to maintenance:
3
• One option is reactive: It is to deal with breakdowns or other problems when they
occur. This is referred to as breakdown maintenance.
• The other option is proactive: It is to reduce breakdowns through a program of
lubrication, adjustment, cleaning, inspection, and replacement of worn parts. This is
referred to as preventive maintenance.
Decision makers try to make a trade-off between these two basic options that will minimize
their combined cost. With no preventive maintenance, breakdown and repair costs would be
tremendous. Furthermore, hidden costs, such as lost production and the cost of wages while
equipment is not in service must be factored in. So must the cost of injuries or damage to
other equipment and facilities or to other units in production. However, beyond a certain
point, the cost of preventive maintenance activities exceeds the benefit.
As an example, if a person never had the oil changed in his or her car, never had it lubricated,
and never had the brakes or tires inspected, but simply had repairs done when absolutely
necessary, preventive costs would be negligible but repair costs could be quite high,
considering the wide range of parts (engine, steering, transmission, tires, brakes, etc.) that
could fail. In addition, property damage and injury costs might be incurred, plus there would
be the uncertainty of when failure might occur (e.g., on the expressway during rush hour, or
late at night). On the other hand, having the oil changed and the car lubricated every morning
would obviously be excessive because automobiles are designed to perform for much longer
periods without oil changes and lubrications.
The best approach is to seek a balance between preventive maintenance and breakdown
maintenance. The same concept applies to maintaining production systems: Strike a balance
between prevention costs and breakdown costs. This concept is illustrated in Figure below.
The age and condition of facilities and equipment, the degree of technology involved, the
type of production process, and similar factors enter into the decision of how much
preventive maintenance is desirable.
Thus, in the example of a new automobile, little preventive maintenance may be needed since
there is slight risk of breakdowns. As the car ages and becomes worn through use, the
desirability of preventive maintenance increases because the risk of breakdowns increases.
Thus, when tires and brakes begin to show signs of wear, they should be replaced before they
fail; dents and scratches should be periodically taken care of before they begin to rust; and the
car should be lubricated and have its oil changed after exposure to high levels of dust and
dirt. Also, inspection and replacement of critical parts that tend to fail suddenly should be
performed before a road trip to avoid disruption of the trip and costly emergency repair bills.
Optimum
Amount of preventive maintenance
3
4.1 Preventive Maintenance
The goal of preventive maintenance is to reduce the incidence of breakdowns or failures in
the plant or equipment to avoid the associated costs. Those costs can include loss of output;
idle workers; schedule disruptions; injuries; damage to other equipment, products, or
facilities; and repairs, which may involve maintaining inventories of spare parts, repair tools
and equipment, and repair specialists.
Some Japanese companies have workers perform preventive maintenance on the machines
they operate, rather than use separate maintenance personnel for that task. Called total
preventive maintenance, this approach is consistent with JIT systems and lean production,
where employees are given greater responsibility for quality, productivity, and the general
functioning of the system.
Unlike preventive maintenance, breakdowns cannot be scheduled but must be dealt with on
an irregular basis (i.e., as they occur). Among the major approaches used to deal with
breakdowns are the following:
i) Standby or backup equipment that can be quickly pressed into service,
ii) Inventories of spare parts that can be installed as needed, thereby avoiding lead times
involved in ordering parts, and buffer inventories, so that other equipment will be less
likely to be affected by short-term downtime of a particular piece of equipment,
iii) Operators who are able to perform at least minor repairs on their equipment, and
3
iv) Repair people who are well trained and readily available to diagnose and correct
problems with equipment.
The degree to which an organization pursues any or all of these approaches depends on how
important a particular piece of equipment is to the overall production system. At one extreme
is equipment that is the focal point of a system (e.g., printing presses for a newspaper, or vital
operating parts of a car, such as brakes, steering, transmission, ignition, and engine). At the
other extreme is equipment that is seldom used because it does not perform an important
function in the system, and equipment for which substitutes are readily available.
The implication is clear: Breakdown programs are most effective when they take into account
the degree of importance a piece of equipment has in the production system, and the ability of
the system to do without it for a period of time.
Finally, forecasts of future demand for the use of the present or new equipment must be taken
into account. The demand for the replacement equipment might differ because of the different
features it has. For instance, demand for output of the current equipment might be two years,
while demand for output of the replacement equipment might be much longer.
These decisions can be fairly complex, involving a number of different factors. On the other
hand, most of us are faced with a similar decision with our personal automobiles:
When is it time for a replacement? (The answer is given in section ………….)
Many mathematical models have been developed to better define and predict aspects of
maintenance. Some of the models available to assist in making decisions concerning product
maintenance are:
(a) Maintenance Model I – This model determines the optimum number of inspections
per facility per unit of time. An inspection is often disruptive, but it usually decreases
downtime because it means fewer breakdowns. The total downtime is expressed by:
TD = xTpf + (kTBf)/x; where,
TD is the total downtime per unit of time for a facility.
x is the number of inspections per facility per unit of time.
k is a constant for a specific facility.
Tpf is the downtime per inspection for a facility.
TBf is the downtime per breakdown for a facility.
By differentiating the equation with respect to x and put it equal to zero, we get:
dTD/dx = Tpf – kTBfx-2 = 0; therefore, x* = [(kTBf)/ Tpf]1/2
3
Where x* is the optimum number of inspections per facility per unit of time.
Therefore the optimum total downtime per unit of time for a facility is given by: T*D
= 2 kTpfTBf
Example:
Assume that the following data are associated with a piece of engineering equipment:
k = 2; Tpf = 0.009 month; TBf = 0.2 month
X* = [2(0.2)/0.009]1/2 = 6.67
The optimum number of inspections per month for the piece of engineering
equipment is 6.67.
(b) Maintenance Model II – This model determines the optimum time interval between
replacements. The goal is to minimize average annual total cost with respect to the
time between replacements or the life of the equipment in years. The average cost
consists of three elements: mean investment cost, mean maintenance cost, and mean
operating cost. The total average cost is
CT = OC1 + MC1 + IC/x + [(x – 1)/2](αoc +αmc); (i) where :
CT is the average total cost.
OC1 is the equipment’s operational cost for the first year.
MC1 is the equipment’s maintenance cost for the first year.
IC is the cost of investment.
X is the equipment life expressed in years.
α oc is the amount by which operational cost increases per year.
α mc is the amount by which maintenance cost increases per year.
Differentiating equation (i) with respect to x and putting it equal to zero, leads to
dCT/dx = (αoc +αmc)/2 – IC/x2 = 0; Therefore, the optimum replacement interval x*
= [2IC/(αoc +αmc)]1/2; substituting in equation (i) gives the minimum average annual
total cost
C*T = OC1 + MC1 - (αoc +αmc)/2 + [2IC(αoc +αmc)]1/2;
(c) Maintenance Model III – This model is concerned with a parallel system composed
of k identical machines or pieces of equipment, with output fed into the next stage of
the production process. For the system success at least one machine must function
normally. Furthermore, the total cost involved in system operation and downtime
losses with respect to k is minimized.
Using queuing theory knowledge, we write the following relationship to obtain the
average proportion of unit of time that the parallel system is unavailable: UA = [λ/
(λ+μ)]k ; where
UA is the system unavailability.
Λ is the constant machine failure.
Μ is the constant machine repair time.
3
Thus the total cost, TC, is given by:
TC = (UA)(DC) + k(MOC) = [λ/(λ+μ)]k(DC) + k(MOC); where
DC is the downtime cost per unit of time.
MOC is the single machine’s operational cost per unit of time.
Differentiating with respect to k and putting it equal to zero, leads to
D(TC)/dk = ([λ/(λ+μ )]kln[λ/(λ+μ)])(DC) + (MOC) = 0;
Thus the optimum number of machines to be used in the parallel configuration for
minimum total cost k* = (ln[-MOC/(DC lnUA1)]/lnUA1);
Where UA1 = λ/(λ+μ)
Example: The following data apply to an engineering system used to produce certain
mechanical parts:
λ = 4 failures per month, μ = 10 repairs per month, MOC = $150, and
DC = $1,500
(d) Maintenance Model IV – This model determines the optimum replacement time for
an item under ordinary periodic replacement policy. Under this policy an item is
replaced with a new one every xp accumulated hours of operation. If the item
malfunctions prior to xp hours, it is repaired only minimally so that its instantaneous
failure rate, λ(x), corresponding to its probability density function, f(x), remains the
same as it was before failure. It is assumed that each failure is detected
instantaneously and the minimum repair time is negligible.
Integrating the above equation over the time interval [0, x], we get
3
F(x) = 0ƒx2x/(20)2 exp[-(x/20)]2 dx = 1- exp[-(x/20)]2
Where F(x) is the cumulative distribution function.
Subtracting from unit:
R(x) = 1-F(x) = 1-(1-exp[-(x/20)]2) = exp[-(x/20)]2
Where R(x) is the reliability at time x.
(e) Maintenance Model V – This model is similar to Model IV except that in this case
the objective is to minimize the total downtime per unit of time – in other words, to
minimize equipment unavailability. The model represents the constant interval
replacement policy. Two important factors associated with this policy are:
• Replacements are carried out at predetermined times irrespective of the age of
the equipment or unit being replaced.
• Replacements are performed when equipment fails.
The optimal value of xp may be obtained in similar fashion to that for Maintenance
Model IV.
3
Ccm=(SOH)(LC)MTTR)/MTBF, where
SOH represents the scheduled operating hours of the equipment
LC is the maintenance labour cost per hour.
MTBF is the mean time between failures, for the equipment.
MTTR is the mean time to repair for the equipment.
Example 2: A heavy-duty motor is scheduled to operate for 3,000 hours annually. The
expected MTBF and MTTR of the motor are 1,000 hours and 10 hours, respectively.
Determine the annual labour cost of corrective maintenance for the motor if the maintenance
labour rate is $25 per hour.
Solution: Ccm=(3000)(25)(10)/1000=$750
The yearly labour cost is $750
(b) Equipment maintenance cost estimation model - This model calculates the cost of
equipment maintenance with the formula: MC=PMC+CMC+SPIC
Where MC is the equipment maintenance cost.
PMC is the cost of preventive maintenance
CMC is the cost of corrective maintenance
SPIC is the cost of spare parts inventory.
Where STpm is the scheduled time preventive maintenance work will take.
TTpm is the expected travel for preventive maintenance
SI pm is the scheduled interval at which preventive maintenance takes place
UH is the number of usage hours, or in use time, per time period considered.
R is the servicing engineer’s hourly rate, including the prorated parts cost.
Example 3: Assume that for maintenance of a personal computer, the following values are
given:
• MTTR = 2 hours, MTBF = 7,500 hours
• UH = 4,500hours per annum
• R = $400 per hour; STpm = 0.35 hour
• ICR = 8% per year; SIpm = 2,500 hours
• OMC = $1,000, TTpm = 0.25 hour; TTcm = 0.25 hour
3
PMC=(0.35+0.25)(4,500)(400)/2,500=$432
CMC=(0.25+2)(4,500)(400)/7,500=$540
MC=432+540+80=$1,052
4.5 Availability
Every designer would like to achieve the highest reliability at minimum cost; however it must
be recognised that equipments will break down sooner or later. When this happens, the speed
and ease with which it can be repaired become vitally important, and hence we must also
consider maintainability.
For any equipment, the probability that it will be available for use is important. Hence,
availability is defined as
Example: An average availability of 0.80 means that the equipment is working satisfactorily
for 80% of the time, and under repair, including waiting for spare, etc for the remaining 20%.
Availability is often more important to the user than reliability. At first sight it might be
thought that a major objective of the maintenance engineer must be to ensure that every item
of equipment is available for use as continuously as possible. This is not necessarily so, for if
it were the case it will make scheduled maintenance useless. The basis of scheduled
maintenance is that item equipment, which is operating acceptably, is deliberately made non-
available while it is checked and any necessary corrections made. The scheduled maintenance
nearly always reduces the average availability. Its attraction, however, is that we can choose a
time when the equipment will not be required for use, whereas breakdowns often occur at
most inconvenient times.
(Breakdowns – whenever a system fails to work as intended i.e. breaks down, the repairs and
adjustments much be made to put it right. Since we can never foresee exactly when such
failure will occur we cannot plan for it, and so this is called unscheduled maintenance as
opposed to scheduled maintenance also referred to as preventive maintenance or planned
maintenance or routine maintenance).
3
(iii) Cost and nuisance value of split, extra setting time, etc.
(iv) Loss of profits due to the system being out of operation, plus the cost of
unrecovered overheads while it is idle.
(v) Damage to good customer relations, which is caused by late deliveries.
4.6.1 Mean Downtime (MDT) and Mean time to repair (MTTR) [or Repair Time]
Downtime (MDT)
Downtime, or outage, is the period during which equipment is in the failed state. Downtime
may commence before repair (e.g. a system not in continuous use may develop a fault while
it is idle; the fault condition may not become evident until the system is required for
operation). Repair often involves an element of checkout or alignment, which may extend
beyond the outage.
4
iii) Alignment time – As a result of inserting a new module into the system
adjustments may be required. As in the case of checkout, some or all of the
alignment may fall outside the downtime.
Activities (i) – (vi) are called Active Repair Elements as opposed to Passive Repair
Activities which consist of:
• Realization time – this is the time, which elapses before the fault condition becomes
apparent. It does not constitute part of repair time;
• Logistic time – this is the time consumed waiting for spares, test gear, additional tools
and manpower to be transported to the system;
• Administrative time – this involves failure reporting, allocation of repair tasks,
manpower changeover due to demarcation arrangement, official breaks, disputes, etc.
Design, maintenance arrangements, environment, and manpower, instructions, tools and test
equipment determine Active Repair Elements. The maintenance environment, that is, the
location of spares, equipment and manpower and the procedure for allocating tasks, mainly
determines logistic and administrative time.
Another parameter related to outage is Repair rate (μ). It is simply the downtime expressed as
a rate, therefore: μ = 1/MTTR
In general, it is the active repair elements that are determined by the design and the passive
elements, which are governed by the maintenance philosophy. Designers must be aware of
the maintenance strategy and of the possible equipment failure modes. Achieving acceptable
repair times involves simplifying diagnosis and repair.
The key design areas are access, adjustment, built-in test equipment, circuit layout and
hardware partitioning, connections, displays and indicators, handling, human and ergonomic
factors, identification, inter-changeability, Least Replaceable Assembly (LRA), mounting,
component selection, redundancy, software, standardization, test points, and, safety (Apart
from legal and ethical considerations, safety-related hazards increase active repair time by
requiring greater care and attention; An unsafe design will encourage short cuts or the
omission of essential activities; Accidents add, very substantially, to the repair time).
Both active and passive repair times are influenced by factors other than equipment design.
Consideration of maintenance procedures, personnel, and spares provisioning is known as
Maintenance Philosophy and plays an important part in determining overall availability. The
costs involved in these activities are considerable and it is therefore important to strike a
balance between over- and under-emphasizing each factor. They can be grouped under seven
headings:
(i) Organization of maintenance of resources
(ii) Maintenance procedures
4
(iii) Tools and test equipment
(iv) Personnel selection, training and motivation
(v) Maintenance instructions and manuals
(vi) Spares provisioning
(vii) Logistics
The level of maintainability of a product determines the kinds of maintenance work that can
and will need to be performed at each point in the product’s life cycle, and the difficulty and
expense of performing them. Maintainability features, such as mean time to repair (MTTR),
therefore influence maintenance costs such as required manpower. For example if the design
calls for the inclusion of built-in test equipment, the time to fault detection and isolation
should be lower. Usually, higher maintainability means less required maintenance, and
therefore lower maintenance costs. In early equipment design, several alternative levels of
built-in test equipment and other factors that can reduce maintenance costs should
considered.
The objective of performing an economic trade-off analysis is to determine all costs for each
alternative under consideration and then to compare them. Usually, the alternative with the
lowest cost should be selected. This approach is also useful in determining whether items
should be designed to be thrown away or to be repaired (cf. integrated circuits).
The factors include the cost of hardware, manpower, training, test equipment and tools, and
repairs facilities, replacement parts, packaging and shipping, repair parts, and supply,
administration, and cataloguing.
The cost of achieving any desired reliability and the subsequent cost of maintenance are
related to each other roughly as shown in the figure below:
minimum
cost
E total F B cost of achieving
Cost C cost reliability
per cost of maintenance
item and repairs
Produced
A
D
Low Reliability Rm High Reliability
Reliability
Fig: The relation between reliability and maintenance costs
At A the reliability is very low and the amount spent on reliability is also low. If we go on
improving reliability we gradually reach the situation where all the obvious things have been
done, and from now on we shall have to spend increasingly more to achieve very little
reliability improvement. If we were unwise enough to demand an impossible reliability of
1.00, costs will sweep away to infinity beyond B.
4
However, when reliability is low, maintenance costs from all the breakdowns are inevitably
high, as shown at C. As the reliability improves so the cost of maintenance falls, until at D, as
reliability approaches 1.00, maintenance costs approach zero.
By adding reliability and maintenance costs we get curve EF, and find that there is a
particular reliability Rm for which the overall cost is a minimum.
Although the above concept is useful, the actual case is nearly always complicated than
suggests since:
i) It does not follow that an improvement in reliability must inevitably cost more.
Better designs, different materials, better quality control during productions, etc.
may achieve improved reliability at little or no extra cost. If scrap is reduced at the
same time, the overall cost may acutely come down (i.e. reliability cost is difficult
to estimate).
ii) Maintenance costs are also difficult to estimate. When an equipment fails, we are
unlikely to be able to foresee associated costs such as:
o The value of production lost through breakdowns, including the cost of the
late deliveries, split batches, idle operators, etc. on the production line.
o The cost of having a piece of equipment out of action. This probably depends
very much on whether it happened to be required for use during the time it
was under repair.
iii) Costs are inextricably mixed up with all the factors related to reliability as
discussed above. We must define precisely what we mean by reliability costs, or
the figures we assign to them will have no meaning.
4
RCM determines the maintenance needs of any facility, system, or equipment in its operating
context. The process, therefore, entails asking questions on the following subjects:
• The functions and related performance standards of the asset in its current
operating context
• Possible ways in which the asset may fail to perform its required functions.
• Causes of each functional failure
• Events that follow each failure
• Significance of each failure
• Measures to prevent failure
• Corrective measures that may be taken if there is no appropriate preventive step
(a) The RCM Process – takes place first during the equipment design and
development phase, when it is used to develop maintenance plans.
During product operation and deployment, these plans are then modified on field
experience. The following two criteria are key to the maintenance plans:
• Parts that are not critical to safety – In this case, preventive maintenance tasks
should be chosen that will decrease the ownership life cycle cost.
• Parts that are critical to safety – In this case, preventive maintenance tasks should
be chosen that will help prevent reliability or safety from dropping to an
unacceptable level, or will help reduce the ownership life cycle costs.
It is through the preventive maintenance program that incipient failures are detected and
corrected, the probability of failure is reduced, hidden failures are detected, and the cost
effectiveness of maintenance program is improved.
4
(c) Perform Fault Tree Analysis
Fault tree Analysis (FTA) defines an undesirable state of the system or product and then
analyses the system or product, in terms of its operation and environment, to determine all
possible ways in which the undesirable event can occur. An FTA is a useful tool to identify
all possible failure causes at all possible levels associated with a system and to identify the
relationship between causes. It can thus improve the design of any specified system, product
or process. An FTA normally takes place during the early design phase and then is
progressively refined and updated as the design develops.
The prerequisites for a fault-tree analysis include clearly defined analysis scope and
objectives, clear identification of assumptions, well defined analysis resolution, thorough
understanding of the system’s design and its operation and maintenance aspects, well-defined
physical bounds and interfaces for the system, a comprehensive review of system operational
experience, and a clear definition of what constitutes system failure or undesirable events.
4
Set and enact the maintenance tasks and their frequencies.
Special instruments are employed for this purpose. Condition monitoring techniques fall into
one of six categories, according to the symptoms or potential failure effects they monitor –
dynamic effects, electrical effects, physical effects, temperature effects, particle effects, and
chemical effects.
(a) Dynamic effects - the methods detect failures, particularly of rotating equipment,
that result in abnormal energy emissions in the form of waves e.g. vibrations, pulses, or noise.
Dynamic monitoring techniques include:
4
problems in Ferro–magnet materials including welds, shafts, boilers, and
machined surfaces.
ii) X-ray radiography – This technique monitors surface and subsurface
discontinuities caused by gas porosity, stress, or fatigue, and also monitors
discontinuities such loose wires in compressors, welds, gearboxes pumps, and
steel structures.
iii) Strain gauges - This technique monitors strain in civil engineering structures such
as tunnels and bridges.
iv) Eddy current testing – This method monitors factors such as material hardness,
and surface and subsurface discontinuities caused by wear, stress, and fatigue in
ferrous materials used in items like heat exchangers tubes, railways lines, boiler
tubes, and hoist ropes.
v) Electron fractography – This technique tracks the growth of fatigue cracks in
motor vehicles, metallic components in air craft, industrial equipment, and similar
items.
vi) Ultrasonic (a pulse echo technique) - This technique is used on welds, boilers,
tubes, compressors, receivers, steel structures, and other items either of ferrous or
non-ferrous materials to monitor the thickness of materials, subject to wear and
corrosion, as well as surface and below-surface discontinuities caused by factors
such as inclusions, fatigue, heat treatment, and lamination.
f) Chemical Effects: some of the techniques used to monitor chemical effects are:
(i) Gas chromatography - This method detects gases emitted as the result of faults
in nuclear power systems and turbine generators.
4
(ii) Thin-larger activation - This technique monitors wear in devices such as
turbine blades, electrical contacts, bearings rails, and cooling systems.
(iii) Infrared spectroscopy - This method measures fluid degradation and the
presence of gases such as hydrogen, carbon monoxide, and methane in
enclosed oil systems such as compressor sumps, Transformers, and engine
sumps.
(iv) Spectrometric oil analysis procedure - This technique is used in circulating oil
systems to track wear, leaks, and corrosion.
(c) Improvement in Teamwork - The practice of RCM not only fosters teamwork within the
RCM review groups but also helps to improve communication and co-operation among
various people and units: design engineers; equipment users, and maintainers; production
and operation units and maintenance a personnel; and management, supervisors; technical
personnel and operators.
(d) Improvement is safety and environmental Rotation - The ways in which the practice
of RCM leads to improved safety and environmental protection include:
(i) The reduction in the total number of frequency of routine tasks automatically
lowers the chance of critical failures occurring either during maintenance or
shortly after equipment start-up.
(ii) The RCM decision process demands that all potential failures that would or could
affect safety or the environment be eliminated or addressed.
4
(iii) Examination of a failure’s safety and environmental implications prior to
considering it, operational effects makes safety and environmental integrity a top
priority.
(iv) The attention to hidden failures and the systematic approach to failure – funding
results in considerable improvement to preventive maintenance. The probability
that multiple failures having serious consequences will occur is thereby
substantially reduced.
(v) Improvement in Individual Motivation – RCM helps improve the motivation of
individuals involved in the review process by providing:
4
The basic tree for LCC starts with a very simple tree based on the costs for acquisition and
the costs for sustaining the acquisition during its life as shown in Figure 1.
Acquisition
Sustaining Costs
Costs
Life Cycle Costs will clearly be reduced by enhanced reliability, maintainability and safety
but will be increased by the activities required to achieve them. In general, cost details for the
acquisition tree shown in Figure 2, are usually identified and collected correctly. The
collection of costs for the sustaining tree shown in Figure 3 is the major problem.
5
Figure 2: Acquisition Cost Tree
5
Figure 3: Sustaining Cost Tree
The operation and maintenance costs of some equipment can be 60 to 80 percent of the life
cycle cost. Life cycle costing is combined with life cycle assessment to consider the costs of
energy consumption and pollution during manufacture and service, and the costs of retiring
the product when it reaches its useful life.
The elements in the life cycle of a product include the overlooked impact on society costs
(OISC) that are rarely quantified and incorporated into a product life cycle analysis. We start
5
with design. The actual costs incurred at the design stage are a small part of the LCC but the
costs committed here comprise about 75 percent of the avoidable costs within the life cycle of
the product. Moreover, it is about 10 times less costly to make a change or correct an error in
design than it is in manufacturing. The costs of acquiring and processing the raw materials
can incur large environmental costs.
The costs of ownership of a product are the traditional aspect of LCC. Useful life is
commonly measured by cycles of operation, length of operation, or shelf life. In design using
durable and reliable materials and components extends the life for use and service of a
product. Maintenance costs, especially maintenance labor costs, usually dominate other use
costs. Most analyses divide maintenance costs into scheduled or preventive maintenance and
unscheduled or corrective maintenance. The mean time between failure (MTBF) and the
mean time to repair (MTTR) influence the LCC. (LCC is reduced by high values of MTBF
and low values of MTTR). Other costs that must be projected for the operations and support
phase are: maintenance of support equipment; maintenance facility costs; pay and fringe
benefits for support personnel; warranty costs and service contracts.
Once the product has reached its useful life it enters the retirement stage of the life cycle.
High-value-added products may be candidates for remanufacturing. By value added we
mean the cost of materials, labor, energy, and manufacturing operations that have gone into
creating the product. Products that lend themselves to recycling are those with an attractive
reclamation value, which is determined by market forces and the ease with which different
materials can be separated from the product. Reuse components are subsystems from a
product that have not spent their useful life and can be reused in another product. Materials
that cannot be reused, remanufactured, or recycled are discarded in an environmentally safe
way. This may require labor and tooling for disassembly or treatment before disposal.
The advantages of life cycle costing are that is an excellent tool for comparing the cost of
competing projects, controlling program costs, selecting among competing contractors,
making decisions associated with equipment replacement, reducing total cost, and conducting
planning and budgeting.
Some of the disadvantages of life cycle costing are that it is time consuming and expensive,
that collecting the data needed for analysis can be a trying task, and that the data available is
sometimes of doubtful accuracy.
5
5.2.4 Why Use LCC?
LCC helps change provincial perspectives for business issues with emphasis on enhancing
economic competitiveness by working for the lowest long term cost of ownership. Too often
parochial views result in ineffective actions best characterized by short term cost advantages
(but long term costly decisions). Consider these typical events observed in most companies:
• Engineering avoids specifying cost effective, redundant equipment needed to
accommodate expected costly failures so as to meet capital budgets,
• Purchasing buys lower grade equipment to get favorable purchase price variances,
• Project engineering builds plants with a 6 month view of successfully running the
plant only during start-ups rather than the long term view of low cost operation,
• Process engineering requires operating equipment in race car driver fashion using a
philosophy that all equipment is capable of operating at 150% of its rated condition
without failure and they have other departments to clean-up equipment abuse,
• Maintenance defers required corrective/preventive actions to reduce budgets, and
thus long term costs increase because of neglect for meeting short term management
gains,
• Reliability engineering is assigned improvement tasks with no budgets for
accomplishing the goals.
Management is responsible for harmonizing these potential conflicts under the banner of
operating for the lowest long term cost of ownership. The glue binding these conflicts
together is a teamwork approach for minimizing LCC. When properly used with good
engineering judgment, LCC provides a rich set of information for making cost effective, long
term decisions. LCC can be used as a management decision tool for:
• Costing discipline - it is concerned with operating and support cost estimates.
• Procurement technique - it is used as a tool to determine cost per usage.
• Acquisition tool - it is concerned with balancing acquisition and ownership costs.
• Design trade-off - it integrates effects of availability, reliability, maintainability,
capability, and system effectiveness into x-y charts that are understandable for cost
effective screening methods.
5
calculating, and understanding reliability principles as improved reliability tends
toward lower life cycle cost.
3. The purpose of maintainability improvement is to design machinery and equipment
that can be quickly and safely repaired to reduce downtime of individual pieces of
equipment so the risk of failing the process is low in both probability of failure and
low in the money risk of exposure. This requires identifying planned maintenance
actions and improving the reliability and longevity of equipment which is driven by
economics. Planning for both quick and safe maintenance activities decreases the
exposure for safety issues and exposure for financial damage from the required
repairs.
Of course, pursing this path for cost reductions has a prerequisite of knowing, understanding,
and using technology from the field of reliability and maintainability. Many papers
concerning reliability and the high cost of unreliability from failures are available for
download at http://www.barringer1.com/Papers.htm .
• Social: Community values and fashion can lead to the need for facility renovation or
replacement, such as environmental and social concerns which give rise to the
obsolescence of processes and products.
• Legal: Revised safety regulations, facility standards, compliance issues or emerging
case law may lead to legal obsolescence.
5
Notwithstanding these six defined life expectancies, and there could be others, the life of an
asset is generally thought to be equal to its economic life. This is the period of time during
which the asset is able to make a positive contribution to the financial position of its owners,
both present and future. The concept of using ‘financial interest’ as the basis for assessing the
time horizon of a LCC study is not unique, but is increasingly giving way to ‘sustainability
considerations’.
Techniques
Given that LCC analysis is most commonly a technique for examining the economic
consequences in the future of choices made now. It may be used to assess the relative
financial merit of a particular proposal or to choose between options. Hence, LCC analysis
will, regardless of the selected model, is typically characterised by:
• a defined time span or investment horizon,
• analysis of economic consequences of current decisions,
• the concept of discount rates to express future and present costs, and
• The use of sensitivity factors.
The central principle of LCC analysis is the value of money over time – a shilling today is
worth more than a shilling in the future, due to its earning potential if invested in the interim
versus the ‘opportunity cost’ of its alternative use.
There are three principle methods of evaluating the LCC of a facility or its components;
1. Current Cost Aggregation: This method simply adds the total capital costs to the
total of all expenditure on operating, repairs, maintenance and replacement over the
facility’s life. This calculation is usually expected to show up in good light proposals
to spend more on the capital cost in order to reduce annual expenditure. However the
‘Simple Aggregation’ has no standing in management accounting terms; this is
because it ignores the highly significant effect of Discounted Cash Flow (DCF) on the
real value of deferred expenditure.
2. Net Present Value: This method, the aggregation of initial and annual expenditure is
modified by deduction of the interest (at the chosen rate) theoretically earned on the
money invested during the period from inception of the project to the actual date of
payment for the facility.
3. Annual Equivalent: This expresses the aggregated amounts in terms of the
‘mortgage payable’ on the initial cost added to the typical annual costs of operating,
repair and replacement. This same quick method which provides a ‘snapshot’ at a
point in time can also be used for comparison of rental options. If considering the
provision for future liabilities this forms the basis of a ‘sinking fund’ formulation.
Each of these methods can be calculated with or without variables such as inflation, tax, etc.
Other important techniques and tools available to support LCC decisions include Internal
Rate of Return (IRR), Break-even Points, Pay-back Periods and Cost Benefit Analysis.
5
(a) General Life Cycle cost estimation models
All these models determine the total cost of an item over its life span, but they vary in the
methods they use to estimate many of the major costs used in the calculation. The following
two cycle cost models demonstrate this point:
(i) Life Cycle cost models I - This model calculates two major kinds of costs:
recurring and nonrecurring. The life cycle cost is expressed by LCC = RC +
NRC
Where LCC is the product’s life cycle cost.
RC is the recurring cost; and NRC is the nonrecurring cost.
The major elements of the recurring cost are operating cost, maintenance cost, support
cost, manpower cost, and inventory cost. For the non recurring cost, they are
procurement cost, reliability and maintainability improvement cost, research and
development cost, installation cost, training cost, support cost, qualification approval
cost, life cycle management cost, test equipment cost, and transportation cost.
The present value of the recurring cost must be obtained, using discounting formulas,
before it is added to the non-recurring cost.
(ii) Life cycle cost model II – Three major costs form the life cycle cost in this
model: Research and development cost, investment cost, and operations and
maintenance cost. Thus the life cycle cost is given by: LCC = RDC + IC +
OMC; where
RDC is the research and Development cost,
IC is the investment cost
OMC is the operations and maintenance cost.
The components of the RDC are engineering design cost, which includes covering system
engineering, reliability, maintainability, human factors, predictability, electrical design,
mechanical design, and logistic support analysis; advanced research and development cost;
engineering development and test cost, for example, the cost of engineering models and of
testing and evaluation, Engineering data cost; and program management cost.
The investment cost consists of construction cost, that is the cost of manufacturing facilities,
test facilities, operational facilities, including the cost of manufacturing engineering, quality
control, fabrication, assembly, tools and test equipment, test and inspection, material, packing
and shipping, initial logistic support cost, or the cost of program management, test and
support equipment, initial spare and repair parts, provisioning initial inventory, first
destination transportation, technical data preparation and initial training and training
equipment.
The elements of the operations and maintenance cost are modification cost, disposal cost,
operations cost, that is, the cost of operations personnel, operational facilities, support and
handling equipment, and operator training, and maintenance cost, including the cost of
maintenance personnel, spare and repair parts, maintenance facilities, maintenance training,
maintenance of test and support equipment, transportation and handling, and technical data.
5
The failure cost FC, is expressed by FC=8T(RC+SC)
Where RC is the cost of repairs
SC is the cost of spares
8 is the constant failure rate of switching power supplies
T is the expected life of the switching power supply.
The cost of spares SC, is given by: SC =(USC) (θ) where θ is the fractional
quantity of spares for each active unit. USC is the unit spare cost.
(ii) Specific life cycle cost Estimation Model II – This model was developed to
estimate the life cycle cost of an early warning radar system. The radar systems
life cycle cost is expressed by: LCCer=Ca+Co+C1
Where LCCer is the life cycle cost of the early warning radar system
Ca is the acquisition cost of the system
Co is the operational cost of the system
C1 is the logistic support cost of the system.
Past experience indicates that Ca accounts for 28% of LCCer, Co for 12% of LCCer, and C1
for 60% of LCCer.
Example 1
Assume that a company is considering buying an electric generator. Manufacturers A, B, and
C are bidding to sell the system. The table below presents data for generators produced by all
three manufacturers. Determine which of the three electric generators has the lowest life
cycle cost.
5
Adding these three costs to the procurement cost, the life cycle cost of the electric generator
from manufacturer A is LCCA=1,500,000+14,430.5+830,372.6+7,265.7 = $2,352,068.8
Adding these costs to the procurement cost, the life cycle cost of the electric generator from
manufacturer B is
LCCB=2,000,000+19,240.6+311.389.7+9,964.4 = $2,340,594.70
The life cycle cost of the electric generator from manufacturer C is:
LCCC=1,800,000+16,835.5+415,186.3+8,563.2 = $2,240,585
Ccm=(SOH)(LC)MTTR)/MTBF, where
SOH represents the scheduled operating hours of the equipment
LC is the maintenance labour cost per hour.
MTBF is the mean time between failures, for the equipment.
MTTR is the mean time to repair for the equipment.
Example 2: A heavy-duty motor is scheduled to operate for 3,000 hours annually. The
expected MTBF and MTTR of the motor are 1,000 hours and 10 hours, respectively.
Determine the annual labour cost of corrective maintenance for the motor if the maintenance
labour rate is $25 per hour.
Solution: Ccm=(3000)(25)(10)/1000=$750
The yearly labour cost is $750
5
(ii) Equipment maintenance cost estimation model - This model calculates the cost
of equipment maintenance with the formula: MC=PMC+CMC+SPIC
Where MC is the equipment maintenance cost.
PMC is the cost of preventive maintenance
CMC is the cost of corrective maintenance
SPIC is the cost of spare parts inventory.
Where STpm is the scheduled time preventive maintenance work will take.
TTpm is the expected travel for preventive maintenance
SI pm is the scheduled interval at which preventive maintenance takes place
UH is the number of usage hours, or in use time, per time period considered.
R is the servicing engineer’s hourly rate, including the prorated parts cost.
Example 3: Assume that for maintenance of a personal computer, the following values are
given:
Solution: SPIC=(1,000)(0.08)=$80
PMC=(0.35+0.25)(4,500)(400)/2,500=$432
CMC=(0.25+2)(4,500)(400)/7,500=$540
MC=432+540+80=$1,052
6
ι ι ) Delivery time – If our deliveries are too long, our customers may buy from a
competitor. However, the time allowed for delivery determines the time available
for design, development, manufacture and test, so if the delivery time is
impossibly short both the quality and reliability of our products may suffer.
ι ι ι ) Quality and performance – we must ensure that we supply to satisfy our
customers. If our quality and performance are too low, our customers may not
buy; if it is unnecessary high, our prices may also be too high.
ι ϖ ) Safety – safety is important and is part of reliability since an equipment which
injures a person is clearly not working as intended.
ϖ) Weight and Volume – Both weight and volume should be reduced to a minimum.
ϖ ι ) Extensibility – the ease with which the equipment can be extended or enlarged
(as the future market requirements increase).
ϖ ι ι ) Adaptability – Adaptability is concerned with the addition of more facilities.
However many facilities, even though they can be added later, are best and
cheapest if included in the new requirement.
Unfortunately the above is quite impossible, because these factors are interrelated. The
greater our demand on quality, reliability, performance, safety, weight and volume, the
greater the cost is likely to be and the longer the delivery time. Therefore, for each individual
case we have to balance one against the other i.e. optimised or trade off one factor against
another.
2. Maintenance is the action taken by the user of an asset to keep it in operable condition
or repair it to operable condition.
6
4. Downtime: This is the total time during which the product is not in an adequate
operating state.
5. Reparability: This is the probability that a failed product will be repaired to its
operational state within a given active repair time.
6. Serviceability: This is the degree of difficulty or ease with which a product can be
restored to its operable state.
7. Accessibility: This is the ease and rapidity with which an item can be reached and the
required maintenance performed. Poor accessibility leads to increased downtime and,
in turn, lower revenue. Thus, one design goal should be to provide access to failed
parts that does not require removing other parts.
9. Visibility: This measures how readily the system part requiring maintenance can
be seen. A blocked view can significantly increase downtime.
10. Testability: This is the measure of fault detection and fault isolation ability. Fault
diagnosis speed can significantly influence downtime and maintenance costs.
11. Active repair time: This is that segment of downtime during which repair staff
work to affect a repair.
12. Logistic time: This is that segment of downtime occupied by the wait for a needed
part or tool.
13. Interchangeability: This is the extent to which one item can be readily replaced
with an identical item without a need for recalibration. Such flexibility in design
reduces maintenance work and in turn maintenance costs.
15. Reliability is the probability that an item will carry out its stated function adequately
for the specified time interval when operated according to the designed conditions.
Reliability at time t = R(t) = (No. Surviving at instant t)/( No. at start when t=o)
16. Failure is the termination of the ability of an item to perform its required function
within the specified guidelines.
17. Hazard rate: This is the rate of change in the number of failed items divided by the
number of items that have not failed at time t.
18. The Mean Time Between Failures (MTBF) tells us how long on average, an
equipment operates before it fails, and this we want to be as long as possible. The
Mean Time To Repair (MTTR) tells us how long on average, it takes to put the
equipment right after it has failed, and this we want to be as short as possible.
19. Availability: This is the probability that a product is available for use when needed.
Average Availability = MTBF/(MTBF + MTTR)
6
20. Utilization factor = MTBF/(MTBF+ MTTR+ MTIBR)
(MTIBR = Mean time the equipment is idle between repairs).
21. In order to achieve reliability, we must have reliable parts/components and use as few
as possible.
22. Design adequacy: This is the probability that the product will complete its
intended mission successfully when it is used according to its design specifications.
At A the reliability is very low and the amount spent on reliability is lso low. At B
reliability is very high and the amount spent on reliability is also high. However,
when reliability is low, maintenance costs from all the breakdowns are inevitably
high, as shown at C. As the reliability improves so the cost of maintenance falls, until
at D, as reliability approaches 1.00, maintenance costs approach zero.
By adding reliability and maintenance costs together we get curve EF, and find that
there is a particular reliability Rm for which the overall cost is a minimum.
24. Redundancy is the provision of more than one means of accomplishing a given
function. There are two types of redundancy – active redundancy and standby
redundancy.
Assume that a machine expected to have a productive life of 10 years is purchased. If at any
time during the ensuring 10 years a new machine is developed that can perform the same task
more efficiently or economically, the old machine has become obsolete. Whether it is “worn
out” or not is irrelevant.
The economic life of a machine is the period over which it provides the best method for
performing its task. When a superior method is developed, the machine has become obsolete.
Thus, the stated book value of a machine can be a meaningless figure.
6
7.2 Depreciation
Whenever any machine or equipment performs useful work its wear and tear is bound to
occur. This can be minimised up to some extent by proper care and maintenance but cannot
be totally eliminated. Its efficiency also reduces with the lapse of time and at one time it
becomes uneconomical to be used further and needs replacements by another new unit.
This reduction of equipment’s efficiency and value with the lapse of time during use is called
Depreciation. Some money, therefore, must be set aside yearly from the profits, so that
when that equipment becomes uneconomical, it can be replaced by the new one. Therefore,
the initial cost of equipment plus installation charges plus repairs charges minus scrap value
is charged against overheads and is spread over equipment’s useful life.
7.3 Obsolescence
Suppose a factory owner purchases a machine for his production shop but after some duration
a better machine comes in the market, whose production rate is very high and much
economical. Although the old machine is efficient but becomes out of fashion and
uneconomical due to the new better machine which has come in the market. This is known as
Obsolescence, and some money should also be set aside from the profits for this cause. Since
it is very difficult to predict whom a new and better machine comes to market, the general
practice is to reduce the life of a machine so as to account the effect of obsolescence. Then
the depreciation and obsolescence charges are calculated in the reduce life.
Example: Suppose the life of a machine is expected to be 20 years then the depreciations
rate will be 100/20 = 5%. By considering obsolescence also, its may be taken as 15 years.
Then the combined depreciation and obsolescence charges will be 100/15 = 6.66% instead of
5%. Therefore, the difference (6.66-5.00) = 1.66% will be obsolescence charges.
Apart from wear and tear the other causes of depreciation are physical decay, accidents,
deferred maintenance and neglect, inadequacy and obsolescence. Depreciation can be broadly
classified as Depreciation due to physical condition and Depreciation due to Functional
condition.
Depreciation
(i) Depreciation due to War and Tear – Whenever any machinery perform work, wear
and tear of certain components take place, although sufficient precautions are taken,
i.e. proper lubricating and cooling is done, which minimises wear and tear but it
6
cannot be totally prevented. Hence, the cost of replacement because of these
precautions is the value of depreciation due to wear and tear.
(ii) Depreciation due to physical decay – There are certain items in a factory, such as
insulation of material, furniture, electric cables, poles, building, chemicals and vessels
etc. which get decay, because of climatic and atmospheric effect, with the result the
value of these articles goes on reducing with the lapse of time. Although every effort
is made by the owner to keep them in serviceable condition, even then because of
climatic and atmospheric effect, there will be reduction in their value. This reduction
in valued is depreciation due to physical decay.
(iii) Accidental Depreciation - Although the machine might have been installed even few
days back and sufficient care is taken to prevent accident, even then, accident may
occur due to some wrong operation, or some loose component or some other cause,
which may result in a heavy damage. The depreciation in machine caused due to this
reason is called accidental depreciation. This knot of depreciation is taken care by
insuring the equipment the amount of the insurance premium depends upon the
estimated cost and life of the equipment.
(iv) Depreciation due to “Deferred Maintenance and Neglect” – If manufacturer’s
instructions for any equipment are neglected and if proper maintenance is not done as
recommended by the manufacturer, then the value of the equipment may be reduced,
and depreciation value because of this, is called depreciation due to “deferred
maintenance and neglect.”
(v) Inadequacy – Inadequacy means reduction in efficiency of an asset. This may result,
even if any equipment is serviced under proper precautions and sufficient
maintenance provided, there is a fall in efficiency with the lapse of time. Secondly,
suppose after say 2-3 years of running, the demand of products manufactured by
certain plant is increase, But the plant cannot cope-up with the increased demand.
This means additional money either to replace with the bigger sized machinery, or
installation of more plants of similar size. This is what is called depreciation due to
inadequacy.
(vi) Depreciation by Obsolescence - Because of technological and scientific
advancement a new machinery comes in the market, which more efficient because of
new invention and better design then, the manufacture of same type of products by the
new machine are much cheaper and better than the existing one, hence the existing
machinery has to be replaced to withstand market competition. This is called
“depreciation by Obsolescence” and is of functional type.
(i) Straight Line Method - This method assumes that the loss of value of machine is
directly proportional to its age. It means one should deduct the scrap value from
the original value and divide the remaining value by the number of years of useful
life.
Let, C be the initial cost plus installation of a machine
S be the scrap value
N be number of years of life of machine, and
D be the depreciation amount per year.
Then D= (C-S)/N
6
This method is also known as “Fixed Instalment” method since same fixed amount is
deducted and no consideration is made about the maintenance and repair charges, which
gradually increases as the machine is getting old.
Example 1: A boiler was purchased for Ksh.900, 000/= on 1st January 1946, the erection and
installation cost was Kshs.140, 000/=. The boiler was to be replaced by a new one on 31st
Dec. 1965. If the scrap value was estimated to be Kshs.300, 000/= what should be the rate of
deprecation and depreciation fund on 15th June, 1955.
Solution:
(a) Total cost = Boiler cost + Erection and Installation charges
(b) As after 12 years tubes have been replaced and cost of replacement is Kshs.30,000/=.
Now book value in 12 years will be Kshs.1,040,000 – 12 x 37,000/= =
Kshs.596,000/= and replacement cost is Kshs.30,000/=. Hence, new book value =
Kshs.596,000 + Kshs.30,000/= = Kshs. 626,000/=.
As, scrap value is same Ksh300,000/= hence the depreciation for the rest 8 years will be
(Kshs.626,000 – Kshs.300,000/=) = Kshs.326,000/=.
:.New rate of depreciation p.a. = Kshs.(326,000/8) = Kshs.40,750/=.
(ii) Diminishing Balance Method (also known as “Reducing Balance “Method) – The
diminishing value of machine is much greater in the early years. It depreciates
rapidly in the early years and later on slowly. Therefore, it is better to depreciate
more during the early years, when the repairs and renewal are not costly.
So, under this method, the book value of the machine goes on decreasing as its
existence continues. A certain percentage of the current book value is taken as the
depreciation.
Let X be the fixed percentage taken to calculate the yearly depreciation on the book
value. Then, X = 1 – (S/C) 1/N
Example 2: A lathe is purchased for Kshs.160,000/= and the assumed life is 10 years, and
scrap value is Kshs.40,000/=. If the depreciation is charged by Diminishing Balance
Method, calculate the percentage by which value of the lathe is reducing every year, and
depreciation fund after 2 years.
Solution: C= 160,000/= ; S = 40,000/= , N = 10
Therefore, X = 1 – (S/N) 1/N = 1- (40,000/160,000)1/10 = 0.1294
Therefore, required % = 12.94
:. Value of the lathe after 1 year = 160,000 (1-0.1294) = Kshs.139,296/=.
:.Depreciation fund after 1 year = Kshs.160,000/= = Kshs.139,296/= = Kshs.20,705/=
Now, value of lathe after 2 years, = Kshs.139,296 (1-0.1294) = Kshs.121,271/=.
:.Depreciation of 2nd year = (139,296 – 121,271) = Kshs.18,025/=
6
:.Depreciation fund after 2 years =20,705 +18,025/= Kshs.38,730/=
(iii) The Insurance Policy Method – This method covers the risk, if the machine becomes
unserviceable before its estimated life.
(iv) Machine Hour Basis Method – In this, rate of depreciation is calculated, considering
the total number of hours a machine runs in a year.
Example 3: The estimated life of a lathe is 10 years and it works 16 hours a day. The initial
cost of lathe is Kshs.160,000/= and scrap value after 10 years is Kshs.50,000/=. If the
machine works for 5,840 hours in a year, calculate the rate of depreciation charged annually
as per machine hour basis method.
Thus suppose the cost of ownership (capital investment) = Kshs.40, 000/= over 10 years.
Total cash inflow = Kshs.60, 000/= over 10 years.
Total profit = Kshs.20, 000/=
:. Average profit per year = Kshs.2, 000/=
:. Percentage return outlay = 2,000/40,000 x 100% =5%
The objection to this method is that it takes no account of when money has to be paid out or
of when it is received.
6
7.6 Installation of new Equipment
Once we have decided to purchase a piece of equipment, we must plan for its installation.
This involves:
(a) The erection of a suitable building to house it.
(b) Preparation of a base or foundations to receive it and the provision of services such as
gas, electricity, water, drainage and compressed air.
(c) Receipt of the equipment and if necessary its storage elsewhere until its permanent
site is ready.
(d) Installation.
(e) Commissioning.
Items (a), (b) and (c) are usually the customer’s responsibility although they may be done to
instructions and drawings supplied by the manufacturer. The latter usually supervises items
(d) and (e).
Equipment replacement decision plays an important role in the economic running of any
organization. There are three types of replacement decisions:
(a) The replacement of capital equipment, as it wears out
(b) The capital equipment required for expansion.
(c) The replacement of equipment due to obsolescence
i.e. the introduction of an improved technology/equipment in the market, which
may produce cheaper products.
Equipments are used to produce at profitable rate, so that the production can stand
competition. Replacement decision is not an easy job, it requires several considerations.
Since it involves large capital investment, a wrong decision may adversely affect the
profitability of the whole concern. Therefore, a scientific approach to solve this problem is
essential. For this purpose the Break-Even Analysis is very useful.
Almost all the equipments are subjected to deterioration and obsolescence in varying degree
with the passage of time. Thus with the passage of time operating inferiority increases. Hence
the old machine has this operating inferiority high and book value as low, while a new
machine to be purchased will have operating inferiority minimum and costs at a maximum.
Hence the problem before the decision maker is to choose between more capital cost and less
imperfection on one hand, and less capital cost and more imperfection on the other.
Although replacement reduces maintenance cost, it involves high capital cost. Therefore an
equipment is replaced when the maintenance and capital cost of the existing equipment is
more than the average capital and operating cost of the new equipment.
(a) Deterioration
It becomes necessary to replace the machine when it wears out and does not function
properly. Such machine starts lowering the quality of products, decreasing the production and
increase in labour and maintenance cost.
(b) Obsolescence
Whenever new equipment (due to new technology) comes in the market, which is capable of
producing more good quality products with less labour and has more efficiency, the existing
6
machine is to be replaced with this machine even though it is functioning well. Generally this
is necessitated because the products manufactured by the new machine will be cheaper.
(c) Inadequacy:
With the change of product design to meet the customers demand or quantity to be
manufactured, old machinery become inadequate and therefore call for different
manufacturing equipment.
When replacing equipment before the expiry of the estimated life, the following reasons or
factors should be considered:
(i) To reduce production cost
(ii) To reduce fatigue
(iii) To raise quality
(iv) To increase output
(v) To secure greater convenience, safety and reliability.
6
(b) Total life average method
(c) Present value method
(d) Rate of return method
(e) MAPI method
This method is not much reliable as it does not take into account its insurance, interest and
maintenance. Further, in the beginning the return is generally less, which increases gradually,
but here we consider it as constant. This method also does not consider depreciation and
obsolescence.
A “present worth money” is today’s value of money invested (at certain interest rate) after
given number of years from today.
Example – If we have Ksh5,000/- to invest and want to know what will be its worth in ten
years at 10% interest, using the formula: F=P(1+i)n, where
F=worth of money in future; P=present amount of money
i=interest rate, n=number of years
Thus Ksh.5,000/- today is worth Ksh12,970/-after ten years from now or in other words
Ksh.12,970/- after ten years from now has a present worth of Ksh5,000/-.
While comparing two alternatives all the costs must be translated into present worth and they
must be compared for equal length of services, e.g. if machine A has its life of 3 years and
machine B has 6 years, then to compare the two machines we must compare the present
worth value of 6 years service of two machines of type A and one machine of type B.
Example: Machine A, operated manually costs, Kshs.2000/= has a life of 2 years. While an
automatic machine B costs Kshs.5000/= but has a life of 4 years. Operating costs for machine
A is Kshs.4,000/= per year while that of machine B is Kshs.3,000/=. Which should be
purchased? Consider 10% interest.
Solution
For comparing the two machines for equal period we must consider 2 machines A against one
machine B. each of the expense is to be converted into present worth.
7
Machine A
Expenses converted in terms of present worth.
(i) Present worth of cost of first piece = Ksh2,000/=
(ii) Present worth of operating cost of first piece.
• in the first year
P = F[1/(1 +i)n] = 4,000/(1 +0.10) = Kshs.3,636/=
• In the second year, P = 4000/(1+0.10)2 = Kshs.3,306/=
(iii) Present worth of cost of second piece of machine A, purchased after the expiry
of life of first piece; i.e. 2 years,
= 2000/ (1 + 0.10)2 = Kshs.1,652/=
(iv) Present worth of operating cost of second piece.
• in the third year = 4000/1+0.10)3 = Kshs.3005/=
• in the fourth year = 4000/(1+0.10)4 = Kshs.2,732/=
Thus the total expenditure in terms of present worth required when machine A is used =
Kshs.16,331/=.
Machine B
Expenses converted in terms of present worth.
(i) Present worth of costs of machine B = Kshs.5000/=
(ii) Present worth of operating cost.
• in the first year = 3000/(1+0.10) = Kshs.2,727/=
• in the second year= 3000/(1+0.10)2 = Kshs.2479/=
• in the third year = 3000/(1+0.10)3 = Kshs.2254/=
• in the third year = 3000/)1+0.10)4 = Kshs.2049/=
But this method has a drawback that earnings of all the years cannot have the value equal to
that of today (present worth). Hence the method will be more useful and practical, if the
earnings, of all years are first converted to present worth and the calculations are made for
rate of return. The following method is an improvement over this method.
C = Investment cost
R = Expected earning in the nth year
R = Rate of return
If we have to find out the rate of return when a machine has worked for n years and earnings
in each year is R1 (in the first year), R2 (in second year), R3 (in the third year) etc, then the
following formula is used (it is calculated for r, by trial and error method).
7
C = R1/(1+r)1 + R2/(1+r)2 + R3/(1+r)3 +…….+Rn/(1+r)n +S/(1+r)n
S = salvage value of the machine after n years.
Example: Find out whether a machine having the following particular must be purchased or
not if rate of interests is 10%.
Since this return of 21% on the investment is higher than the rate of interest of 10%, it is
worthwhile to purchase the machine.
For estimating as to whether the proposed replacement is profitable, the adverse “minimum”
of the defender and the challenger are found and compared. Adverse minimum of the
defender or the challenger is the lowest sum of the time adjusted average of capital cost and
operating inferiority (expressed in terms of money) obtainable from a machine. The
calculations can easily be done with the help of MAPI Charts.
The most important data are failure probabilities and assessments of criticality of the failures,
part failure rates, probability of operator error, and inspection efficiency data. Part failure rate
data can come from experience, banks of generic failure data, and other sources.
The data banks for operator error data fall into three categories: experimentally based data
banks, field based data banks, and subjectively based data banks. The experimentally based
7
data banks contain data gathered in the laboratory. The field based data banks are based upon
data gathered during operation and so often provide more realistic information. The
subjectively based data banks contain data generated by various techniques.
(a) Indicate design and manufacture deficiencies and can be used to support reliability
growth programs; these concern the improvement in reliability, during use, which
comes from field data feedback resulting in modifications. Improvements depend on
ensuring that field data actually lead to design modifications. Reliability growth, then,
is the process of eliminating design-related failures.
(b) Provide quality and reliability trends.
(c) Identify wear-out and decreasing failure rates.
(d) Provide subcontractor ratings.
(e) Contribute statistical data for future reliability and repair time predictions.
(f) Assist second –line maintenance – this for the purpose of:
• Scheduled overhaul and refurbishing of units returned from preventive
maintenance;
• Unscheduled repair and/or overhaul of modules which have failed or become
degraded.
Deeper diagnostic capability is needed and therefore the larger, more complex,
test equipment will be found at the workshop together with full system
information.
(g) Enable spares provisioning to be refined.
(h) Allow routine maintenance intervals to be revised.
(i) Enable the field element of quality costs to be identified.
A failure-reporting system should be established for every project and product. Customer
cooperation with a reporting system is essential if feedback from the field is required (and
this could well be sought, at the contract stage, in return for some concession).
7
• Unit running time.
The main problems associated with failure recording are:
i) Inventories – whilst failure reports identify the numbers and types of failure they
rarely provide a source of information as to the total numbers of the items in question
and their installation dates and running times.
ii) Motivation – if the field personnel can see no purpose in recording information it is
likely that items will be either omitted or incorrectly recorded. If the person is
frustrated by unrealistic time standards, poor working conditions and inadequate
instructions, then the failure report is the first task which will be skimped or omitted.
iii) Verification – Once the failure report has left the person who completes it the
possibility of subsequent checking is remote. If repair times and diagnoses are suspect
then it is likely that they will go undetected or be unverified. Where failure data are
obtained from customer’s staff, the possibility of challenging information becomes
even more remote.
iv) Cost – Failure reporting is costly in terms of both the time to complete failure-report
forms and the hours of interpretation of the information. For this reason, both supplier
and customer are often reluctant to agree to a comprehensive reporting system. (If the
information is correctly interpreted and design or manufacturing action taken to
remove failure sources, then the cost of the activity is likely to be offset by the
savings and the idea must be sold on this basis).
v) Recording non-failures – The situation arises where a failure is recorded although
none exists. This can occur in two ways. First, there is the habit of locating faults by
replacing suspect but not necessarily failed components. When the fault disappears
the first (wrongly removed) component is not replaced and is hence recorded as a
failure. Failure rate data are therefore artificially inflated and spares depleted.
Second, there is the interpretation of secondary failures as primary failures. A failed
component may cause stress conditions upon another, which may, as a result, fail.
Diagnosis may reveal both failures but not always which one occurred first. Again,
failure rates become wrongly inflated. More complex maintenance instructions and
the use of higher-grade personnel will help reduce these problems at a cost.
(vi) Times to Failure: These are necessary in order to establish wear out. In most cases
fault data schemes yield the numbers of failures/defects of equipment. Establishing
the inventories, and the installation dates of items, is also necessary if the cumulative
times are also to be determined. This is not always easy as plant records are often
incomplete (or out of data) and the exact installation date of items has sometimes to
be guessed.
For this to happen it is essential that each item is separately identified (usually by a
tag number) and that each failure is attributed to a specific item.
Further more, if an item is removed, replaced or refurbished as new then this needs to
be identified (by tag number) in order for the correct start times to be identified for
each subsequent failure time.
Another complication is in the use of operating time rather than calendar time. In
some ways the latter is more convenient if the data is to be used for generic use. In
some cases however, especially where the mode is related to wear and the operating
time is short compared with calendar time, then operating hours will be more
7
meaningful. In any case consistency is the rule. If information is available then it is
possible to list:
• Individual times to failure (calendar or operating)
• Times for items which did not fail
• Times for items which were removed without failing
7
(f) Costs of failure such as the penalty cost of system outage (e.g. loss of production)
and the cost of corrective repair effort and associated spares and other
maintenance costs.
(g) The consequences in the case of safety related failures (e.g. death, injury, and
environmental damage) not so easily quantified.
(h) Consideration of whether a failure is intrinsic to the item in question or was
caused by an external factor. External factors might include:
• Process operator error induced failure
• Maintenance error induced failure
• Failure caused by a diagnostic replacement attempt
• Modification induced failure.
(i) Effective data screening to identify and correct errors and to ensure consistency.
There is a cost issue here in that effective data screening requires significant man-
hours to study the field failures returns.
(j) Adequate information about the environment (e.g. weather in the case of
unprotected equipment) and operating conditions (e.g. unusual production
throughout loadings).
It is also useful to know whether the failure rate of a particular failure type is increasing,
decreasing or constant. This will influence the engineering response. A decreasing failure
rate indicates the need for further action in tests to eliminate the failures. Increasing failure
rate shows wear out requiring either a design solution or preventive replacement. Constant
failure rate suggests a reliability level, which is inherent to that design configuration.
7
this sense they yield reliable data. Unfortunately tests tend to be both expensive and time
consuming. Further, although we may know the precise conditions under which a test was
done, we can never be sure that they exactly reproduce the treatment the products will receive
in service. Tests are roughly divided into two:
(i) Demonstration and acceptance tests – These are tests to decide whether a
particular item or items can be accepted.
(ii) Reliability measurements – Tests to estimate values of reliability are likely to be
more expensive than demonstration tests. To measure reliability, a test to estimate
one of the following parameters is carried out, then results converted to reliability
if required:
• The rate of degradation or deterioration
• The MTBF
• The mean time between wear out failures, and the variability about this mean
time.
• The parameters of the Weibull distribution.
Since tests tend to be both expensive and difficult, we should not start one without full
consideration of its probable cost, balanced against the value of the information we hope to
obtain from it.