1

Chapter 1

INTRODUCTION TO
MAINTENANCE AND REPLACEMENT MODELS

Chapter 1
INTRODUCTION TO
MAINTENANCE AND REPLACEMENT MODELS
1.0 MAINTENANCE
Maintenance is a routine and recurring activity of keeping a
particular machine or facility at its normal operating condition so that
it can deliver its expected performance or service without causing any
loose of time on account of accidental damage or breakdown.
Once equipment is designed, fabricated and installed, the
operational

availability

of

the

same

is

looked

after

by

the

maintenance requirement. The idea of maintenance is very old and

was introduced along with inception of the machine. In the early
days, a machine was used as long as it worked. When it stopped
working, it was either repaired/serviced or discarded.
The high cost sophisticated machines need to be properly
maintained/serviced during their entire life cycle for maximizing
their availability. The development of mechanization and automation
of

production

accompanying

systems

and

development

associated
of

ancillary

equipment,
services

with

and

the

safety

requirements, has made it mandatory for engineers to think about

proper maintenance of equipment.
Maintenance is a function to keep the equipment/machine
condition by

replacing or repairing some of the components of the

machine. The maintenance concept is an outline plan of how the

3
maintenance function will be performed. Based on the feedback
obtained from the users and the history of the equipment, detailed
procedures are drawn to concretize the maintenance concept. The
procedures developed thus are collectively called the maintenance
plan. The development of such a maintenance plan is one of the most
important requirements of the maintenance program that requires
interaction between the user and the manufacturer. With this
information, the manufacturer will be in position to rearrange the
design as per users maintenance requirements.
Maintenance function also involves looking after the safety
aspects of certain equipment where the failure of component may
cause a major accident. For example, a poorly maintained pressure
vessel such as steam boiler may cause a serious accident.
1.1 CHALLENGES IN MAINTENANCE:
The maintenance function of a modern industry faces a number of
challenges attributable to:

Rapid growth of technology resulting in current technology

becoming obsolete. Such a challenge is a frequent one in
Information and Communications Technology (ICT) industry where
computers and computers based system (hardware and Software)
is the main component.

Advent of new advanced diagnostic tools, rapid repair systems, etc.

Advance store management techniques to incorporate modular

technologies.

Requirements of keeping both outdated and modern machines in

service. For example, many industrial organizations have a
combination of the old machines working on obsolete technology
and new systems utilizing the latest technology and equipment.
The effective management of maintenance aspects under such

challenging circumstances is often a difficult job. Besides the

rectification of the faults in the equipment, the activities of the
maintenance department include:

Up gradation of the existing plants and equipments and training

maintenance personnel to attend the required technical skills.

Effective maintenance of the old equipment for higher availability

Cost optimization of all maintenance functions

Improvement of maintenance activities in the areas of tribology and

terrotechnology.

Reconditioning of used /unserviceable spare parts.

Development

of

indigenous

sources

for

parts

for

import

substitution.

Setting up of an effective maintenance information management

systems (MIMS).

Effective utilization of the maintenance workforce

Setting up of in house R&D activities for effecting improvements in

maintenance practices.

5
1.2 OBJECTIVIES OF MAINTENANCE:
The objectives of maintenance should be formulated within the
framework of the overall organizational setup so that finally the goals
of the organization are accomplished. For this, the maintenance
division needs to ensure that:
(a) The machinery and/or facilities are always in an optimum working
condition at the lowest possible cost
(b) The time schedule of delivering to the customers is not affected
because of non-availability of machinery/service in working
condition
(c) The performance of the machinery /facility is dependable and
reliable.
(d) The performance of the machinery /facility is kept to minimum to
the event of the breakdown.
(e) The maintenance cost is properly monitored to control overhead
costs.
(f) The life of equipment is prolonged while maintaining the acceptable
level of performance to avoid unnecessary replacements.
Maintenance is also related with profitability through equipment
output and its running cost. Maintenance work enhances the
equipment performance level and its availability in optimum working
condition but adds to its running cost.
The objective of maintenance work should be to strike a balance
between

the

availability

and

the

overall

running

costs.

The

responsibility of the maintenance function should, therefore, be

6
ensure that production equipment /facilities are available for use for
maximum time at minimum cost over a stipulated time period such
that the minimum standard of performance and safety of personal and
machines

are

not

sacrificed.

These

days

therefore,

separate

departments are formed in industrial organizations to look after the

maintenance requirements of equipments and machines.
1.3 TYPES OF MAINTENANCE SYSTEMS
Basically, maintenance can be divided into two groups:
(a) Breakdown maintenance
(b) Planned maintenance
1.3.1 Breakdown Maintenance and Its Limitations
The basic concept of breakdown maintenance is not to do
anything as long as everything is going on well. Hence, no
maintenance or repair work is done until a component or equipment
fails or it cannot perform its normal performance. In other words, the
maintenance work is called upon when the machine is out of order,
and repairs are required to bring back the equipment to its original
working condition.
If the system is alone followed, it will lead to poor operational
availability of the equipment, as spare parts may not be readily
available. Though it appears to be economical proposition, work would
greatly suffer if the machine is not restored to operational condition at
the earliest. In this type of maintenance, during the repair time, no
proper care is taken to know the real cause of the breakdown, which
in turn may lead to frequent failures of the same kind.

7
This type of practice is economical for that machinery whose
breakdown time and repair costs are less. But in case of high cost
production systems, there are several limitations with breakdown
maintenance.
1.3.2 Planned Maintenance:
The planned maintenance is said to be an organized type of
maintenance. In this type of maintenance, the maintenance activities
are planned well in advance to avoid random failure. It will be pre
determined not only the when and what kind of the maintenance
work, but also by whom it would be undertaken. The prerequisites for
planned maintenance include the conduction of work study that
decides the periodicity of maintenance work. Also the conduction of
Time Study helps in suggesting ways and means of devising optimal
maintenance schedules for the given system.
In planned maintenance, instructions will be in greater detail
and specific for each type of equipment. Where safety is of paramount
importance, the equipment condition should be checked everyday.
Hence, the type of maintenance activity to be carried out will depend
upon the nature of equipment and its working conditions.
The planned maintenance can be further classified into:

Scheduled Maintenance (SM)

Preventive Maintenance (PM)

Corrective Maintenance (CM)

Reliability Centered Maintenance (RCM)

8
Scheduled Maintenance:
This is a stitch-in-time procedure to avoid break-downs. The
actual maintenance program is scheduled in consultation with the
production department, so that the relevant equipment is made
available for maintenance work. The frequency of such maintenance
work is decided well in advance from experience so as to utilize the
idle time of the equipment effectively. This also helps the maintenance
department to use their manpower effectively. If the schedule of
maintenance is known in advance, the specialists for the same can
also be made available during the maintenance period. Though
scheduled

maintenance

is

costly

compared

to

breakdown

maintenance, the availability of equipment is enhanced. This practice

is used for overhauling of machines etc.

Preventive Maintenance:
It is said to be preventive maintenance when planned and
coordinated inspections, repairs, adjustments, and replacements are
carried out to minimize the problems of breakdown maintenance. This
is based on the premise that prevention is better than cure. This
practice involves planning and scheduling the maintenance work
without interruption in production schedule and thus improves the
availability of equipment. Under preventive maintenance, a systematic
inspection of each item of equipment or at least the critical parts will
be carried out at predetermined times to unfold the conditions that
lead to production stoppage and harmful depreciation. There is no

9
readymade preventive maintenance plan that suits for any industry. It
should be customized to make it suitable to the requirements of the
particular industry.
Planning and implementation of a preventive maintenance
practice is a costly affair because it involves the replacement of all
deteriorated parts/components during inspection. However, the higher
cost of maintenance usually gets compensated by the prolonged
operational life of the equipment. To avoid serious breakdowns, the
preventive mode of maintenance is usually implemented in complex
plants.

Corrective Maintenance:
The practice of preventive maintenance brings out the nature of
repetitive failures of a certain part of the equipment. When such
repetitive type of failures are observed, corrective maintenance can be
applied so that reoccurrence of such failures can be avoided. These
types of failures can be reported to the manufacturer to suggest
modifications to the equipment.
Corrective maintenance can be defined as the practice carried out
to restore the full performance of the equipment that has stopped
working to acceptable standards. For example, an IC engine may be in
working condition, but does not make its full load because of worn
out piston rings. If the piston rings are replaced, it will bring back the
performance of the engine to specified level.

10
Reliability Centered Maintenance (RCM):
It is used to identify the maintenance requirements of
equipment. The RCM establishes the functional requirements and the
desired performances standards of equipments and these are then
related to design and inherent reliability parameters of the machine.
For each function, the associated functional failure is defined, and the
failure modes and the consequences of the functional failures are
analyzed.
The consequences of each failure are established, which fall in
one

of

the

four

categories:

hidden,

safety

or

environmental,

operational, and no operational. Following the RCM logic, preemptive

maintenance tasks which will prevent these consequences are
selected, provided the applicability and effectiveness criteria for
preventive maintenance are satisfied.
The

applicability

requirements

refer

to

the

technical

characteristics and effectiveness criteria for preventive maintenance

tasks and the frequency at which these should be carried out.
Effectiveness criteria depend on the consequences of the failure;
probabilities of the multiple failures for hidden failure consequences,
acceptable low risk of failure for safety consequences, and nonoperational consequences. When the requirements for planned
maintenance (PM) are not fulfilled, default tasks include failure finding
(for hidden failure, possible redesign of equipment, procedures and
training processes) and no-schedule maintenance.

11
1.4 BENEFITS OF MAINTENANCE:
The high involvement of capital cost in any production system
expects proportional returns from the equipment. These expectations
will be met only when the equipment keeps working at its normal
performance. It is often experienced that the maintenance schedules
provided by the manufacturer do not deliver the required results in
terms of the production out put and the life of the equipment. In such
cases, therefore, it becomes necessary to properly maintain the
equipment with extra care in order to obtain the desired levels of
production or service.

The following benefits can be derived from a well organized

maintenance system:
(a) The minimization of breakdown time
(b) Improvement in total availability of the system with their optimum
capacity
(c) Extended useful life of the equipment
(d) Safety of the personnel.
The consequences of downtime can be very serious when the
machine is working in a production line, as its failure will shut down
the total system. Following a proper maintenance schedule the normal
wear and tear of equipment can be reduced. In certain cases, the
safety of the personnel is of prime importance and this also can be
assured by proper planned preventive maintenance. For example, all

12
aircraft systems need to be inspected before and after a flight as safety
of the passengers is of prime importance.

1.5 EFFECTS OF MAINTENANCE:

Maintenance, being an important function in any production
system, has far reaching effects on the system. If the right practice of
maintenance is not established for a particular environment, it may
lead to serious problem of either over maintenance or under
maintenance. The selection of a particular maintenance policy is also
governed by the past history of the equipment. Cost effective
maintenance will help in enhancing productivity. It is therefore, is
important for the team associated with maintenance work, to know
how much to maintain.
The nature of the maintenance function affects the life of
equipment. It is known from experience that optimum maintenance
will prolong the life of the equipment, and on the other hand,
carelessness in maintenance would lead to reduced life of the
equipment and in some cases an early failure as well. Further, proper
maintenance will help to achieve the production targets. If the
availability of the equipment in good working condition is high, the
reliability of the production will also be high.
Another important effect of the maintenance function is the
working environment. If the equipment is in good working condition,
the operator feels comfortable to use it other wise there is a tendency
to let the equipment deteriorate further. To get the desired results in

13
maintenance operations, there should be selective development of
skilled, semi skilled, and unskilled labour. And also proper job
description is required for the jobs in order to make full use of skilled
workforce available.

1.6 EQUIPMENT MAINTENANCE/ REPLACEMENT/ REENGINEERING:

To decide the effective mode of maintenance it is essential to

carry out reliability analysis of critical parts of the equipment in all
modern automated and semi-automated plants. These critical parts
may be individual pieces of equipment or a combination of parts that
from systems.
Before considering the purchase of any capital equipment,
the evaluation of its reliability is essential, which directly depends
upon the probability of failures. It is desirable to obtain a reliability
index (numerical value) for each machine which is based on such
factors

as

visual

inspection

tests

and

measurements,

age,

environment duty cycle of the equipment. These numbers, so

calculated, represent the reliability of particular equipment. It is also
possible to combine these indices and express an aggregate reliability
index number for the complete system.
From the evaluation of the above index numbers, schedules
can be set for equipment maintenance. Wherever needed, the
maintenance efforts can be expanded. From the reliability reports it is
possible to determine the actions that are required to maintain the
operational availability at the desired level. Cost estimates for such

14
maintenance for much maintenance functions can also be prepared
based on the reliability information.
Similarly, the decision to replace existing equipment will
require the consideration of the following questions, economic factors
and reliability index numbers calculated for the existing equipment.
(a) Will the maintenance cost come down with the replacement of the
old equipment?
(b) Will the cost per unit of production/service come down due to
automated test features of the mew equipment?
(c) Is the existing equipment not sufficient to meet the future
production/service targets?
(d) Will the new equipment be environment friendly and provide better
safety to operators?
(e) Is there any possibility of adding additional accessories to existing
equipment in order to make it more versatile for future use, or is
the rebuilding of existing equipment possible through minor
modifications?

Optimal replacement policy of the equipment can be determined if

reliable estimates of revenue (return from equipment), up keep
(maintenance cost) cost and replacement costs are available. The
equipment in use in industries can be mainly divided into (1)
equipment with diminishing efficiency and (2) equipment with
constant efficiency. The first category deteriorates with time resulting
in increase in operating cost including maintenance cost, and second

15
category operates at constant efficiency for a certain time period and
then deteriorates suddenly.
Several models have been developed using repair vs. time and
cost, in order to solve the replacement problem of equipment with
diminishing

efficiency.

Replacement

is

considered

to

be

the

regeneration point of whole life where the operating cost function

initially starts. In practice such methods really work well and the life
of the equipment/system is enhanced.
On the other hand the concept of reengineering in lieu of
replacement is one viable model as the operating cost increases with
time. This model maximizes the gain between the operating costs
before and after the overhauls. Reengineering can be perceived as the
adjustment, alteration, or partial replacement of a process or product
in order to make it to meet a new need. Successful implementation of
reengineering will improve the equipment or process performance and
this reduces the maintenance and operating costs.

1.7 CONVENTIONAL REPLACEMENT PROBLEM:

The replacement problems are concerned with the issues that
arises when the performance of an item decreases, failure or
breakdown occurs. The decrease in performance or breakdown may be
gradual or sometimes sudden. The need for replacement of items is
felt when,
1. The existing item or system has become inefficient or require more
maintenance.

16
2. The existing equipment has failed due to accident or otherwise and
does not work at all.
3. The existing equipment is expected to fail shortly.
4. The existing equipment has become obsolete due to the availability
of equipment with latest technology and better design.
The solution to replacement problem is nothing but arriving at the
best policy that determines the time at which the replacement is most
economical instead of continuing at an increased maintenance cost.
The Main objective of replacement policy is to direct the organization
in many situations so that it can take right decision. For Example, few
situations are:
(i) Waiting for complete failure of item or to replace earlier at the
expense of higher cost of the item.
(ii) Whether to replace the under performing equipment with the
similar kind of item or by different kind (latest model) of item.
The problem of replacement occurs in the case of both men and
machines. Using probability it is possible to estimate the chance of
death (failure) at various ages.

1.8 TYPES OF FAILURES

As the term failure encompasses wider concept, failures can be
discussed under the following two categories.
(a) Gradual Failure: In this, the failure mechanism is progressive. As
the age of an item increases, its performance deteriorates. This
results in:

17

Increased operating cost

Decreased productivity of the item

Decrease in resale value of item

(Ex: Mechanical items like pistons, bearing rings, tyres, etc.,)

(b) Sudden Failure: This type of failure can be observed in the items
that do not deteriorate gradually with age but which fail suddenly
after some period of service. The time period between installation
and failure will not be constant for any particular equipment.
However

the

failure

pattern

will

follow

certain

frequency

distribution that may be progressive, retrogressive or random in

nature.

Progressive failure:

It is said to be progressive failure, when

probability of failure increases with the age of an item. Ex: light

bulbs, tyres etc.

Retrogressive failure: Certain items will have more probability of

failure in the initial years of their life and with the increase in
the life of an item the chances of failure become less. That is,
the ability of the item to survive in the initial years of life
increases its expected life. Aircraft engines exemplify industrial
equipments with this type of distribution of life span.

Random failure: It is said to be random failure, when constant

probability of failure is associated with equipment that fails
because random causes such as physical shocks that are
independent of age. In the case of random failure, virtually all
items fail before aging has any effect. For example, vacuum

18
tubes, items made of glass or mirror, fruits, vegetables etc may
fail independent of their age.
The replacement situations generally are divided into the following
four types:
(1) Replacement of capital equipment whose performance decreases
with time, e.g., machine tools, vehicles in a transport organization,
airplanes, etc.
(2) Group replacement items that fail completely, e.g., electrical bulbs,
etc.
(3) Problem of mortality and staffing.
(4) Miscellaneous problems.

1.9 REPLACEMENT OF ITEMS THAT DETERIORATE

Determining the optimal replacement period for an item can be
explained by considering an example of a vehicle owner whose aim is
to find the best age at which the old vehicle is to be replaced by a new
one. The vehicle owner intends to ship cargo as cheaply as possible.
The associated costs are:
(i)

The running costs and (ii) the capital cost of the vehicle

These costs can be summarized as average cost per month. It can

be observed that the average monthly cost will go on decreasing, with
increase in time. However, there will be an age at which the rate of
increase in running cost is considerably higher than the savings in
average capital costs. Thus, at this age it is justifiable to replace the
vehicle.

19
1.9.1 Case I: Replacement Policy For Items Whose Maintenance
Cost Increases With Time, And Money Value Does Not Change
With Time i.e. Constant:
Theorem:

The maintenance cost of a machine is given as function

increasing with time and machines scrap value is constant.

(a) When time is a continuous variable, then replacing the machine
when the maintenance cost is equal to the average annual cost will
minimize the average annual cost.
(b) When time is a discrete variable, then replacing the machine when
the maintenance cost in the (n+1)th year becomes greater than the
average annual cost in the nth year will minimize the average
annual cost.
Proof:
(a)

When time t is a continuous variable.

Let Rt =Maintenance cost at time t

C= the capital cost of the item
S= the scrap value of the item
Then the annual cost of the item at any time t = Rt + C - S
The maintenance cost incurred during n years becomes =

dt

The total cost incurred on the item = P (n) =

R dt + C-S
t

Hence average total cost is given by

F(n)=

P ( n)
1
=
n
n

dt +

CS
n

---(1.1)

20
Now, we have to find such time n for which F(n) is minimum.
Therefore, differentiating F(n) with respect to n,
n

dF (n) 1
CS
1
= Rn + 2 Rt dt =0, for minimum of F (n),
dn
n
n2
n 0

---(1.2)

which gives
Rn =

1
n

dt +

CS
P ( n)
=
, by virtue of equation (3.1),
n
n

---(1.3)

Hence, the maintenance cost at time n = average cost in time n

(b)

When time t is a discrete variable

Since the time is measured in discrete units, the cost equation (1.1)
can be written as
F(n)=

P ( n)
=
n

Rt

t 1

CS
n

---(1.4)

By using finite differences, F(n) will be minimum if the following

relationship is satisfied :
F(n-1) < 0 < F(n)

---(1.5)

Now, differencing (3.4) under the summation sign by definition of first

difference,
F(n) = F(n+1) -F(n)
C S n Rt C S
n R
= t
- n
t 1 n 1 n 1 t 1 n

n
R
R
= n 1 t n 1 t 1 n 1

Rt

n
t 1

1
1
+ (C-S)

n 1 n

21
=

Rn 1
n 1

Rn 1
n 1

t 1
n

t 1

1
1
1
1
Rt
+ (C-S)

n 1 n
n 1 n
Rt
CS
n(n 1) n(n 1)

Since F(n)>0 for minimum of F(n), so

Rn 1
>
n 1

Or

t 1

Rt
CS
+
or Rn 1 >
n(n 1)
n(n 1)

t 1

Rt
CS
+
n
n

Rn 1 > P(n)/n, by virtue of equation (1.4)

Similarly, it can be shown that Rn < P(n)/n, by virtue of F(n-1) <0

Hence Rn 1 > ( P(n)/n ) > Rn
This completes the proof.

1.9.2 Case 2: Replacement Policy For Items Whose Maintenance

Cost Increases With Time, And Money Value Changes With
Constant Rate
Money Value:
When it is said that Money is of worth 10% per annum. it can be
interpreted in two ways:
(i) Firstly, investing an amount of Rs.100 today is equivalent to
investing Rs.110 after one year i.e. if we plan to invest Rs.110 after
one year it is equivalent of investing Rs.100 today.
(ii) Secondly, if Rs.100 is borrowed at the rate of interest 10% per
annum and spend this amount today, then it is required to pay
Rs. 110 in a year time.

22
So it can be inferred that Rs.100 in hand today will be equivalent
to Rs.110 in hand after one year from now. In other words, Re.1 in
hand after one year is of worth Rs. 1.1

in hand today.

Present Worth Factor (PWF):

Thus investing Re.1 after one year is equivalent to investing

1.1

rupee today with rate of interest 10% per annum. Re.1 to be

invested after two years from now is equivalent to investing (1.1)2

Rupees today. Therefore Re.1 to be spent after n years is equivalent to
investing (1.1) n rupees today. The quantity (1.1) n is considered as
present worth factor (pwf) or present value of Re.1 to be spent n
years from now.
In general, let money carry an interest rate of r, then 1+r

-n

is

called the present worth factor (pwf) or present value of Re.1 to be

spent after n years from now.

Discount rate (Depreciation value):

Present worth of Re.1 to be spent after n years from now is
given by = 1+r , where r is the interest rate. Then, is called
-n

discount rate or depreciation value.

23
Theorem.: when the maintenance cost increases with time and money
value decreases at constant rate i.e. depreciation value is given, then
the

replacement policy will be: Replace if the next periods

maintenance cost is grater than the weighted average of previous

costs.

Proof:
Let C= Capital cost of the item to be replaced
Ri = Maintenance cost incurred at the starting of the i th year

r = interest rate
= 1+r

-1

is the present worth of Re.1 to be spent a year hence.

The proof can be divided into two major steps:

Step1: To determine the present worth of total cost

Let the item is replaced at the end of every nth year. The year
wise present worth of costs on the item in the successive cycles of n
years can be computed as shown in Table 1.1.
Year
Present
Worth

C+ R1

R2

Rn n1

n+1
(C+ R1 )

n+2

2n

2n+1

R2 n 1

R 2 2n-1

C+R1 2n

Table 1.1: Present worth(Year wise) of future costs

Assuming that the item has no salvage value at the time of
replacement, the present value of all future discounted costs
associated with the policy of replacing the item at the end of every n
year will be given by

24

P(n)= C+R1 + R 2 + + R n n-1 +

C+R 1 n +R 2 n+1+ +R n 2n-1 +

C+R1 2n +R 2 2n+1+ +R n 3n-1 and so on

Summing up the right hand side, we get
P(n) = C+R 1 1+ n + 2n + + R 2 1+ n + 2n + + R n n-1 1+ n + 2n +
= C+R 1 +R 2 + +R n n-1 1+ n + 2n +
= C+R 1 +R 2 + +R n n-1

1
1- n

---(1.6)

[Since <1,the sum of infinite G.P.is

P(n)=

F(n)
,
1- n

P(n+1)=

1
]
1- n

F(n+1)
1- n+1

---(1.7)

where, for simplicity, F(n) = C+ R 1 ++ R n n-1

Step2: To determine replacement policy so that P(n) is minimum.

As n is measured is discrete units, we shall use the method of finite
difference method is used in order to minimize the present worth cost
P (n).

Therefore, if P(n+1) >P(n)> P(n-1), i.e. P(n) > 0 > P(n -1), then P(n)
will be minimum. So by the definition of first difference,
P(n) = P(n+1) P(n) =

F(n+1)
F(n)
from equation (1.7)
n+1
1-
1- n

F(n+1)(1- n )-F(n)(1- n+1)

(1- n+1)(1- n )

Nr

D r form

25
For convenience, we first simplify the N r of P(n) only. That is
N r = F(n+1) (1- n ) F(n)(1- n1 )

= F(n+1) - F(n) + n1 F(n) - n F(n+1)

= R n+1 n + n+1F(n) - n[F(n)+ nR n+1] [ F(n+1) = F(n) + Rn 1 n ]
= v n (1- n )R n+1 - n (1-)F(n)
P(n) =

v n (1- n )R n+1 - n (1-)F(n)

(1- n+1 )(1- n )
(1- n )

n (1-)
R n+1 - F(n)

n+1
n
(1- )(1- ) (1-)

---(1.8)

Simply setting (n-1) for n in equation (1.8)

v n 1 (1 v)
(1 v n )(1 v n 1 )

P(n-1) =

1- n-1

(1-) R n -F(n-1)

After little simplifications of RHS

v n 1 (1 v)
(1 v n )(1 v n 1 )

P(n-1) =

The quantity

1- n

(1-) R n -F(n)

---(1.9)

v n (1 v)
in eqn. (1.8) is always positive,
(1 v n 1 )(1 v n )
since < 1.

Thus, P(n) has the same sign as the quantity under [.] in (1.8), with
similar explanation for P(n-1) in (1.9) also.

Hence the condition, P(n-1) < 0 < P(n), for minimum present worth
expenditure becomes.

1- n
1- n
R n -F(n) < 0 <
R -F(n)
1- n+1
1-

---(1.10)

26

1- n
1- n
R n < F(n) <
R
1- n+1
1-
Rn <

---(1.11)

C+R 1 +R 2+ +R n n-1
<R n+1
1++ 2 + + n-1

Rn <

---(1.12)

F(n)
< R n+1
n-1

---(1.13)

The expression between R n and R n+1 in equations (1.12),and (1.13) is

called the weighted average cost of previous n years with weights
1, , 2 , , n-1respectively.
The value of n satisfying the relationship (1.10), or (1.12) will be the
best replacement age of the item. This proves the theorem.
Annual payment = Weighted average cost for n years.

Selection of best machine:

While making a decision on solution of the best machine,
various costs are to be taken into consideration. The costs that are
constant with time for each given machine are to be taken into
consideration. Some times these costs may differ for each machine.
However those costs that are same for the machines under
comparison may be excluded.
Considering two machines - M1

and

M 2 - selection of an

economically best machine can be done by adopting the following

outlined procedure.

27
Step 1: Find the best replacement age for both machines M1 and M 2
by making use of

Rn <

F(n)
< R n+1
n-1

Let the optimum replacement age for machines M1 and M 2 comes out
be n1 and n 2 , respectively.

Step 2: calculate the fixed annual cost (or weighted average cost) for
each machine by using the formula :
x

C+R 1 +R 2 + +R n n-1
F(n)

2
n-1
1++ + +
n-1

And in this formula substitute n= n1 for machine M1 and n= n 2 for

machine M 2 . Let it be x1 and x 2 for machines M1 and M 2 respectively.
Step 3: (i) If x1 < x 2 , then select machine M1
(ii) If x1 > x 2 , then select machine M 2
(i)

If x1 = x 2 , then both machines are equally good.

1.10 REPLACEMENT OF ITEMS THAT FAIL COMPLETELY

Consider a system usually made up of a large number of low
cost items that are prone to failure with age e.g. failure of a resistor in
television, radio, computer etc. In some cases the failure of a
component may cause the complete failure of the system. In such
cases, the cost of overall failure will be quite higher than the cost of

28
component itself. E.g. the cost of a condenser or tube in an aircraft is
little, but its failure may result in total collapse of the airplane.
When dealing with such situations, two types of replacement policies
shall be considered.
(i)

Individual replacement: In this policy, an item is replaced

immediately after its failure.

(ii)

Group replacement: In this policy, decision is about the age

when all the items should be replaced, irrespective of
whether the items have failed or not. In this policy the items
that fail before the optimal time, will be replaced individually.

1.10.1 Case 1: Individual Replacement Policy

Under this policy an item is immediately replaced after its failure.
To determine the probability of failure (or life span of any item),
mortality tables are used. To discuss such type of replacement policy,
we consider the problem of human population.
Assumptions:
(i) All deaths are immediately replaced by births, and
(ii) There are no other entries or exits.
However in reality it is impossible to have these conditions. But,
the reason for assuming the above two is that the analysis will be
easier by keeping the virtual human population in mind. Such models
can be applied to industrial items, where death of a person is

29
equivalent to the failure of an item or part and birth of a person is
equivalent to replacement. Thus, organizations also face a fairly
common situation. The following Mortality Theorem will make the
conceptions clear.
Mortality Theorem: A large population is subject to a given mortality
law for a very long period of time. All deaths are immediately replaced
by births and there are no other entries or exits. Then the age
distribution ultimately becomes stable and that the number of deaths
per unit time becomes constant ( which is equal to the size of the total
population divided by the mean age at death).
Proof: For convenience, let each death occurs just before some time
t=w, where w is an integer and no member of the population can
survive upto and beyond w+1 time units,
i.e. life span of any member lies between t=0 and t=w.
Let f(t) = number if births at time t,
p(x) = probability of member will die (fail) just before age x+1, i.e.
at age x.
Now f(t-x) = the number of births at time (t-x). The age of such newly
born members who remain alive at time t will obviously be x. This
can be understood from the following Fig. 1.1.
Age

x+1

Time

t-x

t+1

Fig. 1.1: Relation between age and time period

30
Hence, the expected number of deaths of such alive members at time
t is p(x)f(t-x).
Therefore, the total number of deaths at time t will be
w

f (t x) p ( x) , t=w, w+1, w+2,

x 0

Also, total number of births at time (t+1) = f(t+1).

Since all deaths at time t are replaced immediately by births at time
(t+1), therefore
w

f (t 1) f (t x ) p ( x ) ,

t=w,w+1,

---(1.17)

x 0

The difference equation (1.17) in t, may be solved by substituting f(t)

= A t where A is some constant and < 1
Then difference equation (1.17) becomes
w

A t 1 A t x p ( x)
x 0

x 0

x 0

On dividing by A t w , we get w1 w- x p ( x) or w1 w- x p ( x ) 0
Or w1 w p (0) w1 p (1) w 2 p (2) p ( w) 0

--- (1.18)

Since the sum of all probabilities is unity, so

w

x 0

x 0

p( x) 1 or 1 p( x) 0

or

1-[p(0)+p(1)+p(2)++p(w)]=0

---(1.19)

Now comparing equations (1.18) and (1.19), it is found that one

solution of (1.18) is

0 1 . But the polynomial equation (1.18) must

have (w+1) total number of roots. Let the remaining roots be denoted

31
by 1 , 2 , 3 , , w , consequently, the solution of difference equation
(1.17) will be of the form:
f (t ) A0 A11t A2 2t Aw wt

---(1.20)

Where A0 , A1 , , Aw are constants whose value can be determined from

the age distribution at some given point in time. Further, it can be
observed that the absolute value of all the remaining roots is less than
unity i.e.
i < 1 for i= 1,2,3,,w.
Hence, 1t , 2t , , wt tends to zero as t . Consequently, equation
(1.20) becomes f(t)= A0 , which shows that the number of deaths per
unit time ( as well as the number of births) is constant and equal to

A0 .

To show that the age distribution ultimately becomes stable:

Let P(x) =the probability of members remain alive longer than x time
units.
Then, P(x)= 1- P(survivor will die before attaining the age x)
= 1-[p(0)+p(1)++p(x-1)] and P(0)=1

---(1.21)

Since the number of births and deaths have become constant, each
equal to A0 , the expected number of survivors of age x is also stable
at A0 P(x).

32
As the deaths are replaced immediately (i.e. the number of births are
always equal to the number of deaths), the size N of total population
remains constant, i.e.
w

N A0 P ( x) or

---(1.22)

x 0

A0

---(1.23)

P( x)
x 0

Now the number of survivors aged 0,1,2,3, can be calculated from

the equation (1.22) as A0, A0 P (1), A0 P (2), and so on
Finally, if the denominator in (1.23) i.e.

P( x) ,is equivalent to mean

x 0

age at death, then the age distribution will ultimately become stable.
To Prove this,

x 0

x 0

P( x) = P( x) (x)

[ (x)=(x+1)x =1

by finite

differences]
= P ( x ).x 0

w 1

( x 1) p(x)
x 0

=[P(w+1).(w+1) -0] - ( x 1) p(x)

---(1.24)

x 0

But, P(w+1) = 1-p(0) p(1) p(2)--p(w) [from equ.(1.21)]

=0

And

[by virtue of eqn (1.19)]

P(x) =P(X+1) P(x)

= [ 1-p(0) p(1)- p(2)-- p(x)]- [ 1-p(0) p(1)- p(2)-- p(x-1)]
= -p(x)

33
Therefore, substituting the simplified values of P(w+1) and P(x) in
equation (1.24) to obtain.
w

P( x)

= 0+

x 0

w 1

w 1

x 0

y 1

( x 1) p(x) = ( y) p( y 1)

[setting x+1 = y]

y x prob.[that age at death is y]

y 1

= mean (expected) age at death.

The theorem is thus proved.

1.10.2 Case 2: Group Replacement Of Items That Fail Completely

There are certain items viz. Light bulbs that either work or fail
completely. In some cases a system made up of a big number of
similar low cost items that are increasingly prone to failure with age.
While replacing such failed items, always a set-up cost will be there
for replacement. The said set-up cost is independent of the number of
items to be replaced and hence it may be advantageous to replace
entire group of items at fixed intervals. Such a policy is referred as
group replacement policy and found attractive when the value of any
individual item is so less and the cost of keeping records for age of
individual items is not justifiable.

Group Replacement Policy:

Group replacement policy is defined in the following theorem and later
it is explained by numerical example.

34
Theorem:
(a) One should group replace at the end of
individual replacements for the

t th Period if the cost of

t th Period is greater than the

average cost per period through the end of t th period.

(b) One should not group replace at the end of t th Period if the cost of
individual replacement at the end of t th period is less than the
average cost per period through the end of t th period.
Proof:
It is proposed to,
(i) replace all items in group simultaneously at fixed interval t,
whether they have failed or not, and
(ii) continue replacing failed items immediately as and when they fail.
Let N t = number of units failing during time t
N=Total number of units in the system
C(t)=Cost of group replacement after time period t

C1 = Cost of replacing an Individual item

C2 = Cost of replacing an item in group
Then C(t) = C1 [ N1 + N 2 +..+ N t 1 ] + C2 N
Therefore, average cost per unit period will be
F (t )

C (t ) C1[ N1 N 2 N t 1 ] C2 N

t
t

---(1.25)

Now in order to determine the replacement age t, the average cost per
unit period [C(t)/t = F(t), say] should be minimum.
The condition for minimum of F(t) is
F(t-1) < 0 < F(t)

---(1.26)

35
Now, F (t ) F (t 1) F (t )

C (t 1) C (t ) C (t ) C1 N t C (t )

t 1
t
t 1
t

tC1 N t C (t ) C1 N t C (t ) / t

t (t 1)
(t 1)

---(1.27)

which must be greater than zero for minimum F(t), that is

C1 N t >C(t)/t
Similarly, from F(t-1) <0,

---(1.28)

C1 Nt 1 C (t ) / t

---(1.29)

Thus from equations (1.28) and (1.29), the group replacement policy is
completely established.
ILLUSTRATIVE EXAMPLE:
for a certain type of light bulbs( 1000 Nos.), following mortality rates
have been observed:
Week

:1

Percent failing by the end of week

: 10

25

50

80

100

Each bulb costs Rs.10 to replace an individual bulb on failure. If all

bulbs were replaced at the same time in group it would cost Rs. 4 per
bulb. It is under proposal to replace all bulbs at fixed intervals of time,
whether or not the bulbs have burnt out. And also it is to continue
replacing immediately burnt out bulbs. Determine the time interval at
which all the bulbs should be replaced?
SOLUTION:
Let pi = the probability that a new light bulb fails during the i th week
of its life.
Thus p1 = the probability of failure in 1 st week = 10/100 = 0.10
p2 = the probability of failure in 2nd week = (25-10)/100 =0.15

36

p3 = the probability of failure in 3rd Week = (50-25)/100 = 0.25

p4 = the probability of failure in 4 th week = (80-50)/100 = 0.3
p5 = the probability of failure in 5 th week = (100-80)/100 =0.2

Since the sum of all the above probabilities is unity, the further
probabilities p6 , p7 , p8 and so on, will be zero. Thus, all light bulbs are
sure to burnout by the 5th week.
Furthermore, it is assumed that bulbs that fail during a week are
replaced just before the end of that week.
Let N i = the number of replacements made at the end of the i th week.
And let all 1000 bulbs are new initially. Thus,

N0 N0

=1000

N1 N 0 p1 1000 0.10

=100

N 2 N 0 p2 N1 p1 1000 0.15 100 0.10

=160

N 3 N 0 p3 N1 p2 N 2 p1 1000 0.25 100 0.15 160 0.10

=281

N 4 N 0 p4 N1 p3 N 2 p2 N 3 p1

=377

N 5 N 0 p5 N1 p4 N 2 p3 N 3 p2 N 4 p1

=350

N 6 0 N1 p5 N 2 p4 N 3 p3 N 4 p2 N 5 p1

=230

N 7 0 0 N 2 p5 N 3 p4 N 4 p3 N 5 p2 N 6 p1

=286

It has been found that expected number of bulbs failing in each week
increases until 4th week and then decreases until 6th week and again
starts increasing. Thus, the number of failures or replacements will
continue to oscillate till the system settles down to a steady state. In

37
steady state the proportion of bulbs burnt out in each week is
reciprocal of their average life.
Individual replacement:
The mean age of bulbs 1 p1 2 p2 3 p3 4 p4 5 p5
= 1 x 0.1 + 2 x 0.15 +3 x 0.25 +5 x 0.30 +5 x 0.20 = 3.35 Weeks,
The number of failures in each week in steady state =1000/3.35 = 299
And the cost of replacing bulbs individually on failure
=10x299

( at the rate of Rs. 10 per bulb)

= Rs. 2990
Group replacement:
The replacement of all 1000 bulbs at the same time in bulk costs
Rs. 4 per bulb and replacement of an individual bulb on failure costs
Rs. 10. Costs of replacement of all bulbs simultaneously are
calculated in the Table 1.2.
End
of
week
1

Cost of
individual
replacement
100 x 10=1000

Total cost of group

replacement (Rs.)

Average cost
per week (Rs.)

1000 x 4 + 100 x 10=5000

5000.00

160 x 10=1600

5000 + 160 x 10=5000

3300.00

281 x 10=2810

6600 + 281 x 10=9410

3136.67

377 x 10=3770

9410 + 377 x 10=13180

3295.00

Table 1.2: Cost of replacement of bulbs for Bulbs example

The cost of individual replacement in the 4 th week is greater than the
average cost for 3 weeks.
Therefore the optimal replacement decision is to replace all the bulbs
at the end of every 3 weeks. Otherwise the average cost from 4 th week
onwards will start increasing.

38
1.11 EQUIPMENT RENEWAL
The word renewal means either to rope in new equipment in
place of old equipment or repair the old equipment so that the
probability density function of its future lifetime will be equal to that
of new equipment. The probability that a renewal takes place during
the small time interval (t, t+t) is called the renewal rate at time t.
Here time t is measured from the time when the first machine was
started.
The renewal rate of equipment is asymptotically reciprocal of the
mean life of the equipment i.e. h(t )

= reciprocal of mean life

Equipment renewal comes under major preventive maintenance

activity which may include replacement of few parts/subsystems or
conditioning

the

equipment.

With

respect

to

this

detailed

mathematical models are not discussed here as the main focus area is
on blocks and block replacements.
Summary:

The

first

chapter

discusses

the

concepts

and

importance of maintenance in production environment. The two types

of maintenance Breakdown and Planned- are explained. The objective
of maintenance work should be to strike a balance between the
availability and the overall running costs. Also, the possibilities viz.
reengineering the equipment, replacement of the equipment etc. to
ensure the equipment delivers its normal performance, are discussed.
Also this chapter discusses two categories of replacement
techniques for determining the best replacement strategies for the
items that deteriorate with time and those do not deteriorate but fail
suddenly. These models are discussed with respect to the parameters
like maintenance cost, time and value of money.