Professor Dongwook Shin Fall 2021

ISOM2700 Operations Management

Practice Questions Set #4

(Solutions)

1. Assume customers arrive to an M/M/1 queue at the average rate of 10 customers/hour. For safety reasons, the
manager wants to have no more than 2 customers in the system at any time. What should be the service rate to make
sure that the chance of having no more than 2 customers in the system is 90%? Recall that the arrivals are random
so we cannot make 100% sure this never happens.

We know ⋋=10/hour.

P(n≤ 2)=p(n=0)+ p(n=1)+ p(n=2)

10 ! 10 10 " 10 10 # 10
"# ' ( #1 − ' + "# ' ( #1 − ' + "# ' ( #1 − ' = 90%
µ µ µ µ µ µ

10 $
1 − "# ' ( = 90%
µ

μ = 21.5
 ISOM2700 Practice Questions Set 4

2. A cafeteria serving line has a coffee urn from which customers serve themselves. Inter-arrival times at the urn
follow an exponential distribution at the rate of three per minute. In serving themselves, customers take about 15
seconds, exponentially distributed.

a. How many customers would you expect to see on the average at the coffee urn?
b. How long would you expect it to take to get a cup of coffee?
c. What percentage of time is the urn being used?
d. What is the probability that three or more people are in the cafeteria?
e. If the cafeteria installs an automatic vendor that dispenses a cup of coffee at a constant time of 15 seconds,
what how does this change your answers to a and b?

Use M/M/1

l = 3 per minute, µ = 4 per minute

l
a. = L = = 3/(4 – 3) = 3 customers
%& l

(
b. W = l = 3/3 = 1 minute

l
c. ρ = % = 3/4 = .75 or 75%

d. Probability of 3 or more is equal to 1 – probability of 0, 1, 2

$ $ ! $ $ " $ $ #
Po = 41 − )5 #4)5 ' = .2500, P1 = 41 − )5 #4)5 ' = .1875, P2 = 41 − )5 #4)5 ' = .1406

Total of P0 + P1 + P2 = (.2500 + .1875 + .1406) = .5781

Therefore, the probability of three or more is 1 - .5781 = .4219

e. If an automatic vendor is installed, it becomes a G/G/1.

*+!" ,*+#" -./0/12./34 ",! !.78

Wq = 4 5( )𝑊5 = 4 54 5 0.25 = 0.375 minutes
# "&-./0/12./34 # "&!.78

By Little’s Law, Lq = Wq x R = 0.375 x 3 = 1.125 customers

l $
L = Lq + % = 1.125 + ) = 1.875 customers

By converting to constant service time, average waiting time becomes 0.375 minutes, the average time expected to
take a cup of coffee = 0.375 + 0.25 = 0.625 minutes, the average customer in the system becomes 1.875 customers.

2
 ISOM2700 Practice Questions Set 4

3. Mango Electronics Inc. is a Fortune 500 company that develops and markets innovative consumer electronics
products. The development process proceeds as follows.

Mango researches new technologies to address unmet market needs. Patents are filed for products that have the
requisite market potential. Patents are granted for a period of 20 years starting from the date of issue. After receiving
a patent, the patented technologies are then developed into marketable products at five independent development
centers. Each product is only developed at one center. Each center has all the requisite skills to bring any of the
products to market (a center works on one product at a time). On average, Mango files a patent every 7 months (with
standard deviation of 7 months). The average development process lasts 28 months (with standard deviation of 56
months).

a. What is the utilization of Mango’s development facilities?

b. How long does it take an average technology to go from filing a patent to being launched in the market as a
commercial product?

a. We know that l = 1/7 patent per month, μ = 1/28 patent per month, s = 5 servers
Utilization = Flow rate / Capacity = l/sμ = (1/7) / (5*1/28) = .80 or 80%.

CV"! ,CV"# Utilization$2(s+1)-1 W#

b. W9 = # #
'×" 1-Utilization
(×4 s
5

Wq = ((7/7)2+(56/28)2)/2 x (0.8^(SQRT(2*(5+1))-1)/(1-0.8)) x 28 / 5
Wq = 2.5 x 2.885 x 5.6
Wq = 40.4 months

Average wait time = 40.4 months. The average development time = 28 months. Therefore, the total time = 40.4+28
= 68.4 months.

3
 ISOM2700 Practice Questions Set 4

4. Consider a waiting line system with the arrival rate being 10 customers per hour and the service on a customer is
12 minutes on average. Assume that both inter-arrival time and service time follow exponential distribution. Below
is the output from the M/M/s spreadsheet.

1) What’s the arrival rate λ and service rate µ (units per hour)?

λ = 10 customers per hour

µ = 60 minutes / 12 minutes = 5 customers per hour

2) What is the minimum number of servers that are required to ensure the customer’s average waiting time in
queue is no more than 5 minutes?

5 minutes ~ 0.0833 hour

Check the following table, when s = 3, the average waiting time in queue is 0.0889 hour, which is greater than
0.0833 hour. When s = 4, the average waiting time in queue is 0.0174 hour, which is lower than 0.0833 hour.
Therefore, the minimum number of servers is 4.

3) What is the minimum number of servers that are required to ensure the average number of people in the
system is no more than 4?

Check the following table, when s = 2, the average number of people in the system is infinity, which is greater than
4. Shen s = 3, the average number of people in the system is 2.8889, which is smaller than 4. Therefore, the
minimum number of servers is 3.

Inputs:

λ=

µ=

Outputs:

s Lq L Wq W P(0) P(delay) Utilization

1 infinity infinity infinity infinity 0.0000 1.0000 1.0000

2 infinity infinity infinity infinity 0.0000 1.0000 1.0000

3 0.8889 2.8889 0.0889 0.2889 0.1111 0.4444 0.6667

4 0.1739 2.1739 0.0174 0.2174 0.1304 0.1739 0.5000

5 0.0398 2.0398 0.0040 0.2040 0.1343 0.0597 0.4000

6 0.0090 2.0090 0.0009 0.2009 0.1351 0.0180 0.3333

7 0.0019 2.0019 0.0002 0.2002 0.1353 0.0048 0.2857

8 0.0004 2.0004 0.0000 0.2000 0.1353 0.0011 0.2500

9 0.0001 2.0001 0.0000 0.2000 0.1353 0.0002 0.2222

4
 ISOM2700 Practice Questions Set 4

5. The arrival process of customers to a service facility is Poisson with an average arrival rate of 4 customers per
hour. The service facility has a single employee that serves the customers. This person takes on average 10 minutes
to serve one customer (assume this time is exponentially distributed).

(a) What is the service rate?

The service rate µ = 6 customers / hour

(b) What is the average number of customer in the system?

Average queue length in the system L = λ / (µ - λ) = 2 customers.

(c) What is the average time spent by customers in the system?

Average waiting time of customer in the system W = 1 / (µ - λ) = 2/λ = 0.5 hours = 30 minutes per customers

(d) What is the fraction of time that an employee is idle?

The fraction of the time that the employee is idle = 1 – r = 1 / 3 = 33.33% of the time.

5
 ISOM2700 Practice Questions Set 4

6. At a border inspection station, vehicles arrive at the rate of 10 per hour in a Poisson distribution. For simplicity
in this problem, assume that there is only one lane and one inspector, who can inspect vehicles at the rate of 12 per
hour in an exponentially distributed fashion.

(a) What is the utilization of the inspector?

l = 10 per hour µ = 12 per hour

l 10
r= =
µ 12 = .833 or 83.3%

(b) What is the average length of the waiting line?

l2 102
Lq = =
µ(µ - l ) 12(12 - 10) = 4.17 people

(c) What is the average time that a vehicle must wait to get through the system?
: "!
𝐿= = =5
;&: "#&"!

< 8
𝑊= :
= "!
= 0.5 ℎ𝑜𝑢𝑟𝑠 𝑜𝑟 30 𝑚𝑖𝑛𝑢𝑡𝑒𝑠

(d) What is the probability that when you arrive there will be three or more vehicles ahead of you (either in
line or being served)?

Probability of 3 or more is equal to 1 – probability of 0, 1, 2

n
æ l öæ l ö
Pn = çç 1 - ÷÷çç ÷÷
è µ øè µ ø
0 1
æ 10 öæ 10 ö æ 10 öæ 10 ö
P0 = ç 1 - ÷ç ÷ P1 = ç 1 - ÷ç ÷
è 12 øè 12 ø = .1667, è 12 øè 12 ø = .1389

2
æ 10 öæ 10 ö
P2 = ç 1 - ÷ç ÷
è 12 øè 12 ø = .1157

Total of P0 + P1 + P2 = (.1667 + .1389 + .1157) = .4213

Therefore, the probability of three or more is 1 - .4213 = .5787 or 57.87%

6
 ISOM2700 Practice Questions Set 4

7. There are two streams of customers arriving independently at a supermarket: “Express” customers arrive at a
rate 70 per hour, and stay on average 10 minutes within the shop. Each of them spends on average $25. “Regular”
customers arrive at a rate 40 per hour. They stay within the shop for 30 minutes, each spending $80 (on average).

(a) What is the average number of “express” customers in the supermarket, at a given moment of time?

The arrival (flow) rate for express customer is le = 70 customers per hour. Express customers stay on average 10
minutes means We = 10 minutes = 1/6 hour (flow time) for express customers.

Using Little’s law, we have that Le = le ´ We =70 ´ 1/6 = 70/6 =11.67 express customers

(b) What is the average number of “regular” customers in the supermarket, at a given moment of time?

Similarly, for regular customers, the arrival rate is lr = 40 customers per hour and Wr = 30 mins = 1/2 hour.

Using Little’s law, we get Lr = lr ´ Wr = 20 regular customers.

(c) What is the average revenue per hour?

To get the average revenue per hour we need to get the rate at which customers of each time come out of the
supermarket. In general, the rate out of a queueing system is equal to the rate into the system, that is the arrival
rate.

Therefore, average revenue per hour = le ´ $25 + lr´ $80 = 70 x 25 + 40 x 80 = $4950

7
 ISOM2700 Practice Questions Set 4

8. (Optional) Below is the excel simulation of a single-line service system with a single server. Assume the start
time is zero. Fill in the empty cells with numbers.

Cumulative distribution of interarrival times:

From To Interarrival Time

0 0.2 4

0.2 0.5 5

0.5 1 8

Cumulative distribution of service times:

From To Service Time

0 0.3 3

0.3 0.7 6

0.7 1 7

Waiting Service
Interarrival Arrival Service Service
Index Rand() time in Rand() completion
time time start time time
queue time

1 0.9 8 8 0 8 0.5 6 14

2 0.3 5 13 1 14 0.1 3 17

3 0.6 8 21 0 21 0.8 7 28