TUGAS KELOMPOK

SI-5101 ANALISIS REKAYASA

“Waiting Lines and Queuing Theory Models”

Dosen:
Ir. Biemo W. Soemardi, M.SE., Ph.D

Oleh:
Fauziah Shanti Cahyani Siti Maisarah 35018004
Ratih Dewi Shima 25019019
Muhamad Adryan Arif Farrosi 25019032
Ni Luh Ayu Indrayani 25019039
Adlina Kusumadewi 25019041

KELOMPOK KEAHLIAN MANAJEMEN REKAYASA KONSTRUKSI

PROGRAM MAGISTER TEKNIK SIPIL
FAKULTAS TEKNIK SIPIL DAN LINGKUNGAN
INSTITUT TEKNOLOGI BANDUNG
2019
Exercise of Chapter 13:
“Waiting Lines and Queuing Theory Models”

Problems 13-10.
The Schmedley Discount Department Store has approximately 300 customers shopping in its store
between 9 A.M. and 5 P.M. on Saturdays. In deciding how many cash registers to keep open each
Saturday, Schmedley’s manager considers two factors: customer waiting time (and the associated
waiting cost) and the service costs of employing additional checkout clerks. Checkout clerks are paid an
average of $8 per hour. When only one is on duty, the waiting time per customer is about 10 minutes (or
1/6 hour); when two clerks are on duty, the average checkout time is 6 minutes per person; 4 minutes
when three clerks are working; and 3 minutes when four clerks are on duty.

Schmedley’s management has conducted customer satisfaction surveys and has been able to estimate
that the store suffers approximately $10 in lost sales and goodwill for every hour of customer time spent
waiting in checkout lines. Using the information provided, determine the optimal number of clerks to
have on duty each Saturday to minimize the store’s total expected cost.

Answers:

From the information provided above, we can define the data needed, as below.
  Number of Checkout Clerks Working
1 2 3 4
Average number of customers on Saturdays 300 300 300 300
Average time each customer to wait 1/6 1/10 1/15 1/20
Average waiting cost lost sales per hour $10.00 $10.00 $10.00 $10.00
Schemdley's Clerk Salary Cost per hour $8.00 $16.00 $24.00 $32.00

After defining the data, we can calculate the data, shown as table below.
  Number of Checkout Clerks Working
  1 2 3 4
(a) Average number of customers on Saturdays 300 300 300 300
(b) Average time each customer to wait 1/6 1/10 1/15 1/20
(c) Total customer hours lost ((a)x(b)) 50 30 20 15
(d) Average waiting cost lost sales per hour $10.00 $10.00 $10.00 $10.00
(e) Value of the customer's waiting cost ((c)x(d)) $500.00 $300.00 $200.00 $150.00
(f) Schemdley's Clerk Salary Cost per hour $8.00 $16.00 $24.00 $32.00
(g) Schemdley's Clerk Salary Cost in one day (8 hours) $64.00 $128.00 $192.00 $256.00
(h) Total Expected Cost ((e) + (g))   $564.00 $428.00 $392.00 $406.00
As we can see from the table above, the lowest cost of total expected cost is $392 with the optimal
number of total checkout clerks needed on duty of each Saturday are three person. These concluded that
if Scmedley Discount Department Store Manager wants to minimized their expected cost, they have to
put three clerks on Saturday’s duty.

Problems 13-11.
The Rockwell Electronics Corporation retains a service crew to repair machine breakdowns that occur
on an average of λ = 3 per day (approximately Poisson in nature). The crew can service an average of µ
= 8 machines per day, with a repair time distribution that resembles the exponential distribution:

(a) What is the utilization rate of this service system?

(b) What is the average downtime for a machine that is broken?
(c) How many machines are waiting to be serviced at any given time?
(d) What is the probability that more than one machine is in the system? Probability that more than
two are broken and waiting to be repaired or being serviced? More than three? More than four?

Answers:

(a) What is the utilization rate of this service system?

The utilization factor of the machine (ρ, rho), or the probability of the service system being
used,
λ 3
ρ= = =0.375
μ 8
Thus the percentage of time the service crew being busy is 37.5%.

(b) What is the average downtime for a machine that is broken?

Average downtime (W), the time a machine has to wait, plus the time spent by service crew to
repair a machine,
1 1 1
W= = = =0.2 of a day
μ−λ 8−3 5
Assume the service crew works 8 hours per day, so the average downtime for a machine that is
broken is 0.2 x 8 hours = 1.6 hours
(c) How many machines are waiting to be serviced at any given time?
The number of machines that are waiting to be serviced or repaired (L q),

λ2 (3)2
Lq = = =0.225
μ( μ−λ) 8 x (8−3)
Because the result is not integer number, we can assume it is at least 1 machine is waiting to be
serviced at any given time.
 (d) What is the probability that more than one machine is in the system? Probability that
more than two are broken and waiting to be repaired or being serviced? More than three?
More than four?
We can start from the probability when no machine is in the system, given by,
λ 3
P0=1− =1− =0.625
μ 8
Thus, the probability more than 0 machine is in the system is 1 – 0.625 = 0.375. While the
probability that more than one machine, defined as,
k+1 1+1
λ 3
Pn> k =() ()
μ
=
8
=0.141

The formula we use above, can be inputed into a table form, shown as below.

k 3 k+1

()
Pn> k =
8
0 0.375
1 0.141
2 0.053
3 0.020
4 0.007

From the computation table above, we can conclude that the probability more than one is broken
and waiting to be repaired is 14.1%, more than two is 5.3%, more than 3 is 2%, and more than
four is 0.7%.

Problems 13-17.
Automobiles arrive at the drive-through window at a post office at the rate of 4 every 10 minutes. The
average service time is 2 minutes. The Poisson distribution is appropriate for the arrival rate and service
times are exponentially distributed.

(a) What is the average time a car is in the system?

(b) What is the average number of cars in the system?
(c) What is the average time cars spend waiting to receive service?
(d) What is the average number of cars in line behind the customer receiving service?
(e) What is the probability that there are no cars at the window?
(f) What percentage of the time is the postal clerk busy?
(g) What is the probability that there are exactly two cars in the system?

Answers:
From the information above we can define,
Mean number of automobile arrival, λ = 4 cars per 10 minutes = 24 cars per hour,
 10
and the average service time is 2 minutes, thus the µ = = 5 cars per 10 minutes = 30 cars per hour.
2
(a) What is the average time a car is in the system?
The average time a car is in the system (W),
1 1 1
W= = = =0.167 hours=10 minutes
μ−λ 30−24 6
So, the average time of a car spends in the system is 10 minutes.
(b) What is the average number of cars in the system?
The average number of cars in the system (L),
λ 24
L= = =4 cars
μ− λ 30−24
(c) What is the average time cars spend waiting to receive service?
The average time cars spend waiting to receive service (W q),
λ 24
W q= = =0.133 hours=8 minutes
μ ( μ−λ) 30 x( 30−24)
So, the average time cars spend for waiting the service is 8 minutes.
(d) What is the average number of cars in line behind the customer receiving service?
The average number of cars waiting in line behind the current car being serviced (L q),

λ2 ( 24)2
Lq = = =3.2 cars
μ( μ−λ) 30 x (30−24)
Thus, the number of cars waiting behind the current car being serviced is approximately 3 – 4
cars.
(e) What is the probability that there are no cars at the window?
The percentage of idling time, or when there are no cars at the window (P 0),
λ 24
P0=1− =1− =0.2=20 %
μ 30
There are probability about 20% of idle time, no cars at all.
(f) What percentage of the time is the postal clerk busy?
The percentage of the time when the postal clerk is busy (ρ),
λ 24
ρ= = =0.8=80 %
μ 30
There are times when the postal clerk is busy, the chance of them being busy is about 80%.
(g) What is the probability that there are exactly two cars in the system?
The probability that there are exactly two cars (n = 2) in the system, means P n>1 to Pn>2, given by,
k +1 k+1
λ λ
Pn> 1−Pn>2=
μ () () −
μ
1+1 2 +1
24 24
¿ ( ) −( )
30 30
¿ 0.64−0.512
¿ 0.128
 Thus, the probability that there are exactly two cars in the system is 12.8%.

Problems 13-31.
The typical subway station in Washington, D.C., has 6 turnstiles, each of which can be controlled by the
station manager to be used for either entrance or exit control—but never for both. The manager must
decide at different times of the day just how many turnstiles to use for entering passengers and how
many to be set up to allow exiting passengers.

At the Washington College Station, passengers enter the station at a rate of about 84 per minute between
the hours of 7 and 9 A.M. Passengers exiting trains at the stop reach the exit turnstile area at a rate of
about 48 per minute during the same morning rush hours. Each turnstile can allow an average of 30
passengers per minute to enter or exit. Arrival and service times have been thought to follow Poisson
and exponential distributions, respectively. Assume riders form a common queue at both entry and exit
turnstile areas and proceed to the first empty turnstile.

The Washington College Station manager does not want the average passenger at his station to have to
wait in a turnstile line for more than 6 seconds, nor does he want more than 8 people in any queue at any
average time.

(a) How many turnstiles should be opened in each direction every morning?
(b) Discuss the assumptions underlying the solution of this problem using queuing theory.

Answers:

(a) How many turnstiles should be opened in each direction every morning?
From the information provided above, we know that there are 6 turnstiles, with,
Passenger arrival at the station, enter turnstiles, λ = 84 passengers/minute
Passenger arrival from train, exit turnstiles, λ = 48 passengers/minute
Each turnstile can allow, µ = 30 passengers/minute
The manager wants to decide how many turnstiles to use at different times of the day, and how
many turnstiles to be set to allow exiting passengers. While the manager also desired that the
average time the passenger has to wait in line, W q ≤ 6 sec (or 0.1 minute, and no more than 8
people at average time (Lq ≤ 8 passengers).

For computing how many turnstiles should be opened, we can use QM for Windows this time,
by calculated each Wq and Lq for the possibilities any number of turnstiles are opened.

Passangers Enter Turnstiles.

λ = 84 passengers/minute
µ = 30 passengers/minute
For passangers arrival to station entering the turnstiles, we will start from 3 turnstiles opened at
any time, we will not start from 1 turnstile, as this is a multichannel queuing model so it is
required to have the number of server more than 1. We also cannot start from 2, as multichannel
queuing model also required system service rate greater than the arrival rate, where the system
rate is given by the number of each turnstile can allow, multiplied by the number of server
(turnstile). The turnstiles only allow 30 passengers per minute, if it is multiplied by 2 servers
(number of turnstile), it only gives result 60 passangers per minute which no greater than 84
passangers.

If, open 3 turnstile (M=3), the result from QM for windows is shown as below.

Washington College Station Enter Turnstiles

M/M/s by either POM, QM or POM-QM for Windows, V4.

This spreadsheet was created

Data Results
Arrival rate (λ) 84 Average server utilization(r) 0.933
Service rate (µ) 30 Average number of customers in the queue(Lq) 12.273
Average number of customers in the
Number of servers(s) 3 system(L) 15.073
Average waiting time in the queue(Wq) 0.146
Average time in the system(W) 0.179
Probability (% of time) system is empty (P0) 0.016

If, open 4 turnstile (M=4), the result from QM for windows is shown as below.

Washington College Station Enter Turnstiles

M/M/
s
This spreadsheet was created by either POM, QM or POM-QM for Windows, V4.

Data Results
Arrival rate (λ) 84 Average server utilization(r) 0.700
Service rate (µ) 30 Average number of customers in the queue(Lq) 1.000
Number of servers(s) 4 Average number of customers in the system(L) 3.800
Average waiting time in the queue(Wq) 0.012
Average time in the system(W) 0.045
Probability (% of time) system is empty (P0) 0.050
If, open 5 turnstile (M=5), the result from QM for windows is shown as below.

Washington College Station Enter Turnstiles

M/M/s
This spreadsheet was created by either POM, QM or POM-QM for Windows, V4.

Data Results
Arrival rate (λ) 84 Average server utilization(r) 0.560
Service rate (µ) 30 Average number of customers in the queue(Lq) 0.241
Number of servers(s) 5 Average number of customers in the system(L) 3.041
Average waiting time in the queue(Wq) 0.003
Average time in the system(W) 0.036
Probability (% of time) system is empty (P0) 0.058

The summary from three result above, is shown as table below.

M Wq Lq Manager Requirement
3 0.146 12.273 Not okay, Because Lq > 8 person, and Wq > 0.1
4 0.012 1 Okay
5 0.003 0.241 Okay

Hence, the opened turnstiles for passengers which entering the station is either 4 or 5 turnstiles.
But if we choose 5 turnstiles, it will only 1 turnstiles left for the exiting passengers. Thus, we
should compute for the number of turnstiles needed by exiting passangers before conclude the
final number of which turnstile shall be opened.

Passengers Exit Turnstiles.

λ = 48 passengers/minute
µ = 30 passengers/minute

With QM for Windows, we shall start from number of server (turnstile) opened are 2, with the
similar reasons as explanation before. The only difference is that we now have passengers
arrival 48, thus if we multiplied the average number of passengers served at one minute with
two servers, the result will be greater than 48.
If, open 2 turnstile (M=2), the result from QM for windows is shown as below.

Washington College Station Exit Turnstiles

M/M/s
This spreadsheet was created by either POM, QM or POM-QM for Windows, V4.

Data Results
Arrival rate (l) 48 Average server utilization(r) 0.800
Service rate (m) 30 Average number of customers in the queue(Lq) 2.844
Number of servers(s) 2 Average number of customers in the system(L) 4.444
Average waiting time in the queue(Wq) 0.059
Average time in the system(W) 0.093
Probability (% of time) system is empty (P0) 0.111

If, open 3 turnstile (M=2), the result from QM for windows is shown as below.

Washington College Station Exit Turnstiles

M/M/s
This spreadsheet was created by either POM, QM or POM-QM for Windows, V4.

Data Results
Arrival rate (l) 48 Average server utilization(r) 0.533
Service rate (m) 30 Average number of customers in the queue(Lq) 0.313
Number of servers(s) 3 Average number of customers in the system(L) 1.913
Average waiting time in the queue(Wq) 0.007
Average time in the system(W) 0.040
Probability (% of time) system is empty (P0) 0.187

The summary from three result above, is shown as table below.

M Wq Lq Manager Requirement
2 0.059 2.844 Okay
3 0.007 0.313 Okay

From the summarize above, we can see both are okay, but, if we take three turnstile to be
opened for the exiting passangers, which means also open three turnstiles for enter passanger,
then the turnstiles for entering passengers are overcrowded and does not meet the manager
requirement.

Therefore, the ideal choice is 4 turnstiles opened for entering passangers, and 2 turnstiles
opened for exiting passengers.
(b) Discuss the assumptions underlying the solution of this problem using queuing theory.
In the information provided, it says, “Assume riders form a common queue at both entry and
exit turnstile areas and proceed to the first empty turnstile”. It may be reasonable to assume that
arrivals at the entrance turnstiles are independent and Poisson. But we cannot be sure if the
exiting passengers are really independent case. Moreover, they arrive in batches when the train
is arriving. Unless trains unload every minute or two, this assumption may be unreasonable.
Other problems arise as well. If an exiting passenger’s card does not have the correct fare, the
card is rejected and the passenger must leave the line, go to an “add fare” machine to correct the
deficiency, and enter the queue again. This resembles the reneging customer.