Queueing theory and Replacement model

1. Trucks at a single platform weigh-bridge arrive according to Poisson probability distribution. The time
required to weigh the truck follows an exponential probability distribution. The mean arrival rate is 12
trucks per day and mean service rate is 18 trucks per day. Determine

(i) What is the probability that no trucks are in the system?

(ii) What is the average number of trucks waiting for service?

(iii) What is the average time a truck waits for weighing service to begin?

(iv) What is the probability that an arriving truck will have to wait for service?

Solution:

Solution:

𝛿 = 12 𝑡𝑟𝑢𝑐𝑘𝑠 𝑝𝑒𝑟 𝑑𝑎𝑦, 𝜇 = 18 𝑡𝑟𝑢𝑐𝑘𝑠 𝑝𝑒𝑟 𝑑𝑎𝑦

𝛿 12
𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 𝑓𝑎𝑐𝑡𝑜𝑟 𝜙 = = = 0.67
𝜇 18

The probability that no trucks are in the system is P0 = 1 − ϕ = 1 − 0.67 = 0.33

Average number of trucks in the system

𝜙 0.67
𝐿𝑠 = = = 2.03
1 − 𝜙 1 − 0.67

The average number of trucks waiting for service

δ
Lq = Ls − = 2.03 − 0.67 = 1.36
μ

The average time a truck waits for weighing service to begin

𝐿𝑞 1.36
𝑊𝑞 = = = 0.113 𝑑𝑎𝑦𝑠
𝛿 12
The probability that an arriving truck will have to wait for service

𝛿 12
𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 𝑓𝑎𝑐𝑡𝑜𝑟 𝜙 = = = 0.67
𝜇 18

2. The city council of a small town has decided to build a tennis court in the central park. Players are
expected to arrive on the average of 10 sets of players per 12 hour a day. Playing time is exponentially
distributed with a mean of 1 hour. Arrivals are Poisson. What are the expected queueing statistics
assuming the basic single server model?
Solution:

10
𝛿= 𝑝𝑒𝑟 𝑕𝑜𝑢𝑟, 𝜇 = 1 𝑝𝑒𝑟 𝑕𝑜𝑢𝑟
12
𝛿 10
𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 𝑓𝑎𝑐𝑡𝑜𝑟 𝜙 = = = 0.83
𝜇 12

Average number of players in the park

10
𝜙 12 0.83
𝐿𝑠 = = = = 4.88
1−𝜙 1− 10 1 − 0.83
12
Average number of players in the queue

δ
Lq = Ls − = 4.88 − 0.83 = 4.05
μ

Average waiting time of players in the park

𝐿𝑠 4.88
𝑊𝑠 = = = 0.49 𝑕𝑜𝑢𝑟𝑠
𝛿 10

Average waiting time of players in the queue

𝐿𝑞 4.05
𝑊𝑞 = = = 0.41 𝑕𝑜𝑢𝑟𝑠
𝛿 10
3. Customers arrive at a one window drive at a bank according to Poisson distribution with mean 10 per
hour. Service time per customer is exponential with mean 5 minutes. The space in front of the window
including that for the serviced car can accommodate a maximum of 3 cars. Others can wait outside this
space. Determine

(i) The probability that an arriving customer can drive directly to the space in front of the window.

(ii) The probability that an arriving customer will have to wait outside the indicated space.

(iii) The length of time an arriving customer is exposed to wait before starting service.

Solution:
60
δ = 10/hour, 𝜇 = 5
= 12/𝑕𝑜𝑢𝑟

The probability that an arriving customer can drive directly to the space in front of the window is
𝑛
𝛿 𝛿
𝑃 𝑥 < 3 = 𝑃 𝑥 = 0 + 𝑃 𝑥 = 1 + 𝑃 𝑥 = 2 𝑤𝑕𝑒𝑟𝑒 𝑃𝑛 = 𝜙𝑛 1−𝜙 = 1−
𝜇 𝜇
 0 1 2
𝛿 𝛿 𝛿 𝛿 𝛿 𝛿
= 1− + 1− + 1−
𝜇 𝜇 𝜇 𝜇 𝜇 𝜇
0 1 2 2
𝛿 𝛿 𝛿 𝛿 𝛿 𝛿 𝛿 𝛿 𝛿
= 1− + 1− + 1− = 1− 1+ +
𝜇 𝜇 𝜇 𝜇 𝜇 𝜇 𝜇 𝜇 𝜇
2
10 10 10 2 10 100 2
= 1− 1+ + = 1+ + = 2.53 = 0.42
12 12 12 12 12 144 12

The probability that an arriving customer will have to wait outside the indicated space is
3 𝑘
𝛿 𝛿
𝑃 𝑛≥3 = 𝑤𝑕𝑒𝑟𝑒 𝑃 𝑛 ≥ 𝑘 =
𝜇 𝜇
3
10
𝑃 𝑛≥3 = = 0.58
12

The length of time an arriving customer is exposed to wait before starting service is

𝐿𝑞
𝑊𝑞 =
𝛿
10
𝜙 0.83
𝐿𝑠 = = 12 = = 4.88
1 − 𝜙 1 − 10 1 − 0.83
12
δ
Lq = Ls − = 4.88 − 0.83 = 4.05
μ

𝐿𝑞 4.05
𝑊𝑞 = = = 0.41 𝑕𝑜𝑢𝑟𝑠
𝛿 10
4. Arrivals at telephone booth are considered to be Poisson with an average time of 10 minutes between
one arrival and the next. The length of the phone call is assumed to be distributed exponentially, with
mean 3 minutes.

(i) What is the probability that a person arriving at the booth will have to wait?

(ii) The telephone department will install a second booth when convinced that an arrival would expect
waiting for at least 3 minutes for a phone call. By how much should the flow of arrivals increase in order
to justify a second booth?

(iii) What is the average length of the square that forms from time to time?

Solution:

1 1
𝛿= /𝑚𝑖𝑛𝑢𝑡𝑒𝑠, 𝜇 = /𝑚𝑖𝑛𝑢𝑡𝑒𝑠
10 3
The probability that a person arriving at the booth will have to wait

1
𝛿 10
𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 𝑓𝑎𝑐𝑡𝑜𝑟 𝜙 = = = 0.3
𝜇 1
3

The second booth will be installed, if the new waiting time Wq′ is greater than 3. Let the corresponding
arrival rate be 𝜆1 .

Wq′ > 3

𝛿1 𝛿1
>3⇒ >3
𝜇 𝜇 − 𝛿1 1 1
3 3 − 𝛿1

1
⇒ 𝛿1 > 1 − 3𝛿1 ⇒ 4𝛿1 > 1 ⇒ 𝛿1 >
4
1 1 3
𝐼𝑛𝑐𝑟𝑒𝑎𝑠𝑒 𝑖𝑛 𝑎𝑟𝑟𝑖𝑣𝑎𝑙𝑠 𝑖𝑠 𝛿1 − 𝛿 = − =
4 10 20
Average length of the square that forms from time to time

1 1
μ 3 10
Lw = = = 3 = = 1.43
μ−δ 1− 1 7 7
3 10 30
5. There are two clerks in a university to receive fees from the students. If the service time for each
student is exponential with mean 4 minutes and if the boys arrive in a Poisson fashion at the counter at
the ratio of 10 per hour, determine

(i) The probability of having to wait for service.

(ii) The expected percentage idle time for each clerk.

Solution:

1 60
𝐶 = 2, 𝛿 = 10 𝑝𝑒𝑟 𝑕𝑜𝑢𝑟, 𝜇 = 𝑝𝑒𝑟 𝑚𝑖𝑛𝑢𝑡𝑒𝑠 = = 15 𝑝𝑒𝑟 𝑕𝑜𝑢𝑟
4 4
𝛿 10 10 1
𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 𝑓𝑎𝑐𝑡𝑜𝑟 𝜙 = = = =
𝜇𝐶 2 15 30 3
−1
𝑐−1 𝛿 𝑛 𝛿 𝑐
𝜇 𝜇 𝑐𝜇
𝑃0 = + .
𝑛! 𝑐! 𝑐𝜇 − 𝛿
𝑛=0
 𝑛 −1
2−1 10 10 2
15 2 15
= + 15 .
𝑛! 2! 2 15 − 10
𝑛=0

0 1 −1
10 10 10 2
15 2 15
= + 15 + 15 .
0! 1! 2! 2 15 − 10

−1 −1
0.45 30 0.45 30 −1
= 1 + 0.67 + . = 1.67 + . = 1.67 + 0.34
2! 30 − 10 2! 20
−1
= 20.1 = 0.5

The probability of having to wait for service

𝛿 𝑐
𝜇. 𝜇
𝑃 𝑛≥𝑐 = 𝑃
𝑐 − 1 ! 𝑐𝜇 − 𝛿 0

10 2
15. 15
= 0.5 = 0.167
2 − 1 ! 2(15) − 10

The expected percentage idle time for each clerk is

1 2
= 1 − 𝜙 × 100 = 1 − × 100 = × 100 = 0.67 × 100 = 67%
3 3

Replacement Problems
Replacement of items whose maintenance and repair costs increase with time and ignoring
changes in the value of the money during the period

1. The cost of machine is Rs. 6100 and its scrap value is 100. The maintenance cost found from
experience is as follows:

Year: 1 2 3 4 5 6 7 8
Maintenance cost (Rs.): 100 250 400 600 900 1200 1600 2000
When should the machine be replaced?

Solution:
 Capital Scrap Maintenance Average annual cost
Years Total cost
cost value cost 𝒇(𝒕) 𝟏
n 𝑪 − 𝑺 + 𝒇(𝒕) 𝑪 − 𝑺 + 𝒇(𝒕)
C S 𝒇(𝒕) 𝒏
1 6100 100 100 100 6100 6100
2 6100 100 250 350 6350 3175
3 6100 100 400 750 6750 2250
4 6100 100 600 1350 7350 1837.5
5 6100 100 900 2250 8250 1650
6 6100 100 1200 3450 9450 1575
7 6100 100 1600 5050 11050 1578.571
8 6100 100 2000 7050 13050 1631.25
Here the minimum average annual cost is Rs. 1575. Therefore the machine can be replaced at the end of
sixth year.

2. Machine A cost Rs. 45000 and the operating costs are estimated at Rs. 1000 for the first year
increasing by Rs. 10000 per year in the second and subsequent years. Machine B cost Rs. 50000 and
operating cost are Rs 2000 for the first year, increasing by Rs. 4000 in the second and subsequent years.
If we now have a machine of type A, should we replace it with B? If so when? Assume that both
machines have no resale value and future costs are not discounted.

Solution:

Machine A:

Capital Scrap Maintenanc Average annual cost

Years Total cost
cost value e cost 𝒇(𝒕) 𝟏
n 𝑪 − 𝑺 + 𝒇(𝒕) 𝑪 − 𝑺 + 𝒇(𝒕)
C S 𝒇(𝒕) 𝒏
1 45000 0 1000 1000 46000 46000
2 45000 0 11000 12000 57000 28500
3 45000 0 21000 33000 78000 26000
4 45000 0 31000 64000 109000 27250
Here the minimum average annual cost of Machine A is Rs. 26000. Therefore the Machine A can be
replaced at the end of third year.

Machine B:

Capital Scrap Maintenance Average annual cost

Years Total cost
cost value cost 𝒇(𝒕) 𝟏
n 𝑪 − 𝑺 + 𝒇(𝒕) 𝑪− 𝑺 + 𝒇(𝒕)
C S 𝒇(𝒕) 𝒏
1 50000 0 2000 2000 52000 52000
2 50000 0 6000 8000 58000 29000
3 50000 0 10000 18000 68000 22666.67
4 50000 0 14000 32000 82000 20500
5 50000 0 18000 50000 100000 20000
6 50000 0 22000 72000 122000 20333.33
Here the minimum average annual cost of Machine B is Rs. 20000. Therefore the Machine B can be
replaced at the end of third year.

Total cost of Machine A for the first year = 𝑅𝑠. 46000

Total cost of Machine A for the second year = 𝑅𝑠. 57000 – 46000 = 𝑅𝑠. 11000 < 20000

Total cost of Machine A for the third year = 𝑅𝑠. 78000 – 57000 = 𝑅𝑠. 21000 > 20000

∴ The Total cost of Machine A for the third year is greater than the minimum average annual cost of
Machine B.

∴ Machine A is replaced by Machine B at the end of second year.

Replacement of items whose maintenance costs increases with time and value of the money
also changes with Time

1. A manufacturer is offered two machines A and B. A has cost price of Rs. 2,500, its running cost is Rs.
400 for each of first years and increased by Rs. 100 every subsequent year, Machine B having the same
capacity as A, costs Rs. 1250 and has running cost of Rs. 600 for 6 years, increasing by Rs. 100 per year
thereafter. Taking money’s value as 10% per year, which machine should be purchased? Scrap value of
both the machines is negligibly small.
Solution:
𝑟 = 10% = 0.1

1 1
𝑣= = = 0.9091
1 + 𝑟 1 + 0.1

Machine A

Capital cost C= Rs. 2500, Scrap Value S= 0,

Discounted
Running Discount 𝑪+
Years running 𝑪+ 𝑹𝒏 𝒗𝒏−𝟏
𝑪 cost factor 𝒗𝒏−𝟏
n cost 𝑹𝒏 𝒗𝒏−𝟏 𝒗𝒏−𝟏
𝑹𝒏 𝒗𝒏−𝟏
𝑹𝒏 𝒗𝒏−𝟏
1 2500 400 1 400 2900 1 2900
2 2500 400 0.9091 363.64 3263.64 1.9091 1709.52
3 2500 400 0.8265 330.6 3594.24 2.7356 1313.88
4 2500 400 0.7513 300.52 3894.76 3.4869 1116.97
5 2500 400 0.683 273.2 4167.96 4.1699 999.53
6 2500 500 0.621 310.5 4478.46 4.7909 934.78
7 2500 600 0.5645 338.7 4817.16 5.3554 899.5
8 2500 700 0.5132 359.24 5176.4 5.8686 882.05
9 2500 800 0.4665 373.2 5549.6 6.3351 876.01
10 2500 900 0.4241 381.69 5931.29 6.7592 877.51
Hence machine A should be replaced after 9th year.

Machine B:

Capital cost C= Rs. 1250, Scrap Value S= 0

Discounted
Running Discount 𝑪+
Years running 𝑪+ 𝑹𝒏 𝒗𝒏−𝟏
𝑪 cost factor 𝒗𝒏−𝟏
n cost 𝑹𝒏 𝒗𝒏−𝟏 𝒗𝒏−𝟏
𝑹𝒏 𝒗𝒏−𝟏
𝑹𝒏 𝒗𝒏−𝟏
1 1250 600 1 600 1850 1 1850
2 1250 600 0.9091 545.46 2395.46 1.9091 1254.76
3 1250 600 0.8265 495.9 2891.36 2.7356 1056.94
4 1250 600 0.7513 450.78 3342.14 3.4869 958.48
5 1250 600 0.683 409.8 3751.94 4.1699 899.77
6 1250 600 0.621 372.6 4124.54 4.7909 860.91
7 1250 700 0.5645 395.15 4519.69 5.3554 843.95
8 1250 800 0.5132 410.56 4930.25 5.8686 840.11
9 1250 900 0.4665 419.85 5350.1 6.3351 844.52

Hence machine A should be replaced after 8th year.

Since the weighted average cost in 9 years of machine A is Rs. 876 and weighted average cost in 8 years
of machine B is Rs. 840, it is advisable to purchase machine B.

Replacement of items that fail suddenly

1. A plant has 6 numbers of cells and a study has been made on the nature of deterioration with time.
The following is the probability of failure after replacement.

Months after replacement 1 2 3 4 5

Probability of failure 0.2 0.15 0.1 0.3 0.25
The cost of replacement as and when it fails has been found to be Rs. 200 per unit. Now it has been
proposed to replace all the items at a specified frequency and in such a case the cost of such a
replacement comes to Rs. 60 per unit. Can we accept this proposal? If so, at what frequency we have to
undertake replacement of all the cells at a time?

Solution:

Let 𝑃𝑖 be the probability that a cell fails during the ith month of its life.

Let 𝑁𝑖 represent the number of replacements made at the end of ith month when all 6 cells are new
initially. Then we have
Month Expected number of failures
0 𝑁0 = 6
1 𝑁1 = 𝑁0 𝑃1 = 6 × 0.2 = 1.2
2 𝑁2 = 𝑁0 𝑃2 + 𝑁1 𝑃1 = 6 × 0.15 + 1.2 × 0.2 = 1.14
3 𝑁3 = 𝑁0 𝑃3 + 𝑁1 𝑃2 + 𝑁2 𝑃1 = 6 × 0.1 + 1.2 × 0.15 + 1.14 × 0.2 = 1.008
𝑁4 = 𝑁0 𝑃4 + 𝑁1 𝑃3 + 𝑁2 𝑃2 + 𝑁3 𝑃1
4 = 2.2926
= 6 × 0.3 + 1.2 × 0.1 + 1.14 × 0.15 + 1.008 × 0.2
𝑁4 = 𝑁0 𝑃5 + 𝑁1 𝑃4 + 𝑁2 𝑃3 + 𝑁3 𝑃2 + 𝑁4 𝑃1
5 = 2.5837
= 6 × 0.25 + 1.2 × 0.3 + 1.14 × 0.1 + 1.008 × 0.15 + 2.2926× 0.2

Individual replacement policy:

𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑙𝑖𝑓𝑒 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 = 𝑖𝑃𝑖 = 1 × 0.2 + 2 × 0.15 + 3 × 0.1 + 4 × 0.3 + 5 × 0.25 = 3.25
𝑖

𝑇𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 6

𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑎𝑖𝑙𝑢𝑟𝑒𝑠 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕 = = = 1.85 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕
𝑎𝑒𝑟𝑎𝑔𝑒 𝑙𝑖𝑓𝑒 𝑜𝑓 𝑐𝑒𝑙𝑙𝑠 3.25

𝐶𝑜𝑠𝑡 𝑜𝑓 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙 𝑟𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡

= 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑎𝑖𝑙𝑢𝑟𝑒𝑠 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕 × 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙 𝑟𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡 𝑐𝑜𝑠𝑡 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕

= 1.85 × 200 = 𝑅𝑠. 370 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕

Group replacement policy:

Average
End of
Total cost of replacing cells cost per
month
(T) month
(n)
T/n
1 6 × 60 + 1.2 × 200 = 600 600
2 6 × 60 + 1.2 + 1.14 × 200 = 812 406
3 6 × 60 + 1.2 + 1.14 + 1.008 × 200 = 1029.6 343.2
4 6 × 60 + 1.2 + 1.14 + 1.008+ 2.2926 × 200 = 1488.12 372
5 6 × 60 + 1.2 + 1.14 + 1.008+ 2.2926+ 2.5837 × 200 = 2004.86 401

Minimum average cost per month is 343. So group replacement can be made at the end of third month.

Group replacement period = 3 months

Individual replacement cost/month = Rs. 370

Minimum group replacement cost/month = Rs. 343

Since the minimum group replacement cost/month is lesser than the individual replacement
cost/month, the group replacement policy is best and hence all the cells are to be replaced once in three
months and the cells which fail during this three month period are to be replaced individually.

2. The probability 𝑃𝑛 of failure just before age n (in months) of 1000 light bulbs is given. If individual
replacements costs Rs. 12.5 and group replacement costs Rs. 3.0 per bulb, find the optimal replacement
policy.

n 1 2 3 4 5
𝑃𝑛 0.1 0.2 0.25 0.3 0.15
Solution:

Let 𝑃𝑖 be the probability that a light bulb fails during the ith month of its life.

Let 𝑁𝑖 represent the number of replacements made at the end of ith month when all 1000 cells are new
initially. Then we have

Month Expected number of failures

0 𝑁0 = 1000
1 𝑁1 = 𝑁0 𝑃1 = 1000 × 0.1 = 100
2 𝑁2 = 𝑁0 𝑃2 + 𝑁1 𝑃1 = 1000 × 0.2 + 100 × 0.1 = 210
3 𝑁3 = 𝑁0 𝑃3 + 𝑁1 𝑃2 + 𝑁2 𝑃1 = 1000 × 0.25 + 100 × 0.2 + 210 × 0.1 = 291
𝑁4 = 𝑁0 𝑃4 + 𝑁1 𝑃3 + 𝑁2 𝑃2 + 𝑁3 𝑃1
4 = 396
= 1000 × 0.3 + 100 × 0.25 + 210 × 0.2 + 291 × 0.1
𝑁4 = 𝑁0 𝑃5 + 𝑁1 𝑃4 + 𝑁2 𝑃3 + 𝑁3 𝑃2 + 𝑁4 𝑃1
5 = 330
= 1000 × 0.15 + 100 × 0.3 + 210 × 0.25 + 291 × 0.2 + 396 × 0.1

Individual replacement policy:

𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑙𝑖𝑓𝑒 𝑜𝑓 𝑙𝑖𝑔𝑕𝑡 𝑏𝑢𝑙𝑏𝑠 = 𝑖𝑃𝑖 = 1 × 0.1 + 2 × 0.2 + 3 × 0.25 + 4 × 0.3 + 5 × 0.15 = 3.2
𝑖

1000
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑎𝑖𝑙𝑢𝑟𝑒𝑠 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕 = ≅ 313
3.2
𝐶𝑜𝑠𝑡 𝑜𝑓 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙 𝑟𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡

= 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑎𝑖𝑙𝑢𝑟𝑒𝑠 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕 × 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙 𝑟𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡 𝑐𝑜𝑠𝑡 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕

= 313 × 12.5 ≅ 𝑅𝑠. 3913 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕

Group replacement policy:

 Average
End of
Total cost of replacing cells cost per
month
(T) month
(n)
T/n
1 1000 × 3 + 100 × 12.5 = 4250 4250
2 1000 × 3 + 100 + 210 × 12.5 = 6875 3438
3 1000 × 3 + 100 + 210 + 291 × 12.5 = 10513 3504
4 1000 × 3 + 100 + 210 + 291 + 396 × 12.5 = 15463 3866
5 1000 × 3 + 100 + 210 + 291 + 396 + 330 × 12.5 = 19588 3918

Minimum average cost per month is 3438. So group replacement can be made at the end of second
month.

Group replacement period = 2 months

Individual replacement cost/month = Rs. 3913

Minimum group replacement cost/month = Rs. 3438

Since the minimum group replacement cost/month is lesser than the individual replacement
cost/month, the group replacement policy is best and hence all the tube lights are to be replaced once
in two months and the cells which fail during this two month period are to be replaced individually.

3. The probability 𝑃𝑛 of failure just before age n shown below for 1000 bulbs. If the individual
replacements costs Re. 1 and the group replacement costs Rs. 0.3 per item, find the optimal
replacement policy.

n 1 2 3 4 5
𝑃𝑛 0.3 0.1 0.1 0.2 0.3
Solution:

Let 𝑃𝑖 be the probability that a bulbs fails during the ith month of its life.

Let 𝑁𝑖 represent the number of replacements made at the end of ith month when all 1000 cells are new
initially. Then we have

Month Expected number of failures

0 𝑁0 = 1000
1 𝑁1 = 𝑁0 𝑃1 = 1000 × 0.3 = 300
2 𝑁2 = 𝑁0 𝑃2 + 𝑁1 𝑃1 = 1000 × 0.1 + 300 × 0.3 = 190
3 𝑁3 = 𝑁0 𝑃3 + 𝑁1 𝑃2 + 𝑁2 𝑃1 = 1000 × 0.1 + 300 × 0.1 + 190 × 0.3 = 187
𝑁4 = 𝑁0 𝑃4 + 𝑁1 𝑃3 + 𝑁2 𝑃2 + 𝑁3 𝑃1
4 = 305
= 1000 × 0.2 + 300 × 0.1 + 190 × 0.1 + 187 × 0.3
𝑁4 = 𝑁0 𝑃5 + 𝑁1 𝑃4 + 𝑁2 𝑃3 + 𝑁3 𝑃2 + 𝑁4 𝑃1
5 = 489
= 1000 × 0.3 + 300 × 0.2 + 190 × 0.1 + 187 × 0.1 + 305 × 0.3
Individual replacement policy:

𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑙𝑖𝑓𝑒 𝑜𝑓 𝑏𝑢𝑙𝑏𝑠 = 𝑖𝑃𝑖 = 1 × 0.3 + 2 × 0.1 + 3 × 0.1 + 4 × 0.2 + 5 × 0.3 = 3.1
𝑖

1000
𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑎𝑖𝑙𝑢𝑟𝑒𝑠 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕 = ≅ 323
3.1

𝐶𝑜𝑠𝑡 𝑜𝑓 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙 𝑟𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡

= 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑓𝑎𝑖𝑙𝑢𝑟𝑒𝑠 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕 × 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙 𝑟𝑒𝑝𝑙𝑎𝑐𝑒𝑚𝑒𝑛𝑡 𝑐𝑜𝑠𝑡 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕

= 323 × 1 ≅ 𝑅𝑠. 323 𝑝𝑒𝑟 𝑚𝑜𝑛𝑡𝑕

Group replacement policy:

Average
End of
Total cost of replacing cells cost per
month
(T) month
(n)
T/n
1 1000 × 0.3 + 300 × 1 = 600 600
2 1000 × 0.3 + 300 + 190 × 1 = 790 395
3 1000 × 0.3 + 300 + 190 + 187 × 1 = 977 326
4 1000 × 0.3 + 300 + 190 + 187 + 305 × 1 = 1282 321
5 1000 × 0.3 + 300 + 190 + 187 + 305 + 489 × 1 = 1771 354

Minimum average cost per month is 321. So group replacement can be made at the end of fourth
month.

Group replacement period = 4 months

Individual replacement cost/month = Rs. 323

Minimum group replacement cost/month = Rs. 321

Since the minimum group replacement cost/month is lesser than the individual replacement
cost/month, the group replacement policy is best and hence all the bulbs are to be replaced once in four
months and the cells which fail during this four month period are to be replaced individually.