CB2203 Data-Driven Business Modeling

Tutorial 4
Simulation by Hand

Question 1

Susan runs a pizza restaurant and each day she needs to decide how many orders of
pizza ingredients to purchase from the wholesaler. The number of pizza sales each
day is between 1 to 7, and the frequency distribution of pizza sales (X) is obtained
using sales from previous 100 days:
Pizza Sales (X) 1 2 3 4 5 6 7
Number of 5 12 16 25 20 16 6
days

In this question, you will design a SIMULATION BY HAND experiment over 10 days
(i.e. 10 trials). The random numbers from 00-99 have already been generated in
List 1 shown on the next page.

1. Calculate the cumulative probability based on historical data, assign the

corresponding 2-digit random numbers and then fill in the following form.

Pizza Sales Probability Cumulative Random Number

Probability Assignment
(previous (must start with
cumulative prob 01)
+ current prob.)
1 5% 0.05 5% 01-05
2 12% 0.12 17% (5%+12%) 06-17
3 16% 0.16 33% (17%+16%) 18-33
4 25% 0.25 58% (33%+25%) 34-58
5 20% 0.2 78%(58%+20%) 59-78
6 16% 0.16 94%(78%+16%) 79-94
7 6% 0.06 100%(94%+6%) 1 95-00
Total 100% 1 100% 1

2. Using simulation by hand, run a simulation with 10 trials/days to determine

the number of pizza sales per day. Start from the top of random number List
1. What is the average pizza sales based on the simulation results?
Day Random Number # of pizza sales
1 21 3
2 85 6
3 71 5
4 48 4
5 39 4
6 31 3
7 35 4
8 12 2
9 73 5
10 41 4
Average number of pizza sales: (3+6+5+4+4+3+4+2+5+4)/10=4

1
CB2203 Data-Driven Business Modeling

3. Using an analytical technique, what is the theoretical average sales (or

expected sales) per day? (refering to Q1 table, use the prob, X use
cumulative prob))
Expected sales= all possible value of X* corresponding probability
= number of pizza sales (X)* probability
=1*0.05+2*0.12+3*0.16+4*0.25+5*0.2+6*0.16+7*0.06
= 4.15

4. Suppose Susan orders 5 pizza ingredients on each day and makes all the
ingredients into pizza at the beginning of the day (i.e., Pizza supply is fixed at
5 each day). Each pizza is sold for $10, and unsold pizza will be donated to
local charity and receive $0. The cost of each pizza is $5. How would you
calculate the average profit over the 10-day period simulated in question 2?
Notes: Unsold pizza will bring $0 revenue. In days of pizza shortages,
customers will just switch to other pizza stores for purchasing pizza.
[Profit = $10*max(5,0) - 5*5]
Day Random # of pizza Cost Revenue Profit
Number sales ($5/pizza) ($10/pizza) (Cost-Revenue)
1 21 3 $5*5=25 3*$10=30 30-25=5
2 85 6 25 5*$10=50 50-25=25
3 71 5 25 5*$10=50 50-25=25
4 48 4 25 4*$10=40 40-25=15
5 39 4 25 4*$10=40 40-25=15
6 31 3 25 3*$10=30 30-25=5
7 35 4 25 4*$10=40 40-25=15
8 12 2 25 2*$10=20 20-25= -5
9 73 5 25 5*$10=50 50-25=25
10 41 4 25 4*$10=40 40-25=15
Average profit: sum(all profit column)/total number of tdy
(5+25+25+15+15+5+15-5+25+15)/10
= $14

2
CB2203 Data-Driven Business Modeling

List 1 – Random numbers to determine the number of pizza sales. Start

from the top of the list. There may be more random numbers than you
need.
21
85
71
48
39
31
35
12
73
41
31
97
78
94
66
74
90
95
29
72

3
CB2203 Data-Driven Business Modeling

Question 2 (similar to asg)

The manager of a bank in North Carolina, is attempting to determine how many
tellers are needed at the drive-in window (like counter) during peak times. As a
general policy, the manager wishes to offer service such that average customer
waiting time does not exceed 2 minutes (criterion of mangement).

Data for Service Time

Service Time Probability
(Minutes) (Frequency)
0 0
1.0 0.25
2.0 0.2
3.0 0.4
4.0 0.15

Data for Customer Arrivals

Time between Successive Probability
Customer Arrivals (time (Frequency)
between 2 customers = time
interval)
0 0.10
1.0 0.35
2.0 0.25
3.0 0.15
4.0 0.10
5.0 0.05

List 2 – Random numbers to determine the time of customer arrival. Start

from the top of the list. There may be more random numbers than you
need.
50
28
68
36
90
62
27
50
18
36
61
21
46
01
14
81
87
72
80
46

4
 CB2203 Data-Driven Business Modeling

List 3 – Random numbers to determine the service time. Start from the top
of the list. There may be more random numbers than you need.
52
37
82
69
98
96
33
50
88
90
50
27
45
81
66
74
30
59
67
60

(a) Calculate the cumulative probability based on above data tables (Service Time
& Customer Arrivals), assign the corresponding 2-digit random numbers and
then fill in the following two forms.
Data for Service Time
Service Time Probability Cumulative Random Number
(Minutes) (Frequency) Probability Assignment
0 0.00 0.00 (impossible)
1.0 0.25 0.25 01-25
2.0 0.20 0.45 26-45
3.0 0.40 0.85 46-85
4.0 0.15 1 89-00
Data for Customer Arrivals
Time between Probability Cumulative Random Number
Successive (Frequency) Probability Assignment
Customer Arrivals
0 0.10 0.1 01-10
1.0 0.35 0.45 11-45
2.0 0.25 0.7 46-70
3.0 0.15 0.85 71-85
4.0 0.10 0.95 86-95
5.0 0.05 1 96-00

(b) Assume the bank will open at 9:00 am and there is only one teller available, fill
in the following form to design a SIMULATION BY HAND experiment over 15
customers (i.e. 15 trials).
Custome Rando Interval Time of Random Servic Start End Wait Idle

5
 CB2203 Data-Driven Business Modeling

r m To Arrival Number e Servic Service Time Time

Number Numbe Arrival Time e
r
1 50 2.0 9:02 52 3.0 9:02 9:05 0 2
2 28 1.0 9:03 37 2.0 9:05 9:07 2 0
3 68 2.0 9:05 82 3.0 9:07 9:10 2 0
4 36 1.0 9:06 69 3.0 9:10 9:13 4 0
5 90 4.0 9:10 98 4.0 9:13 9:17 3 0
6 62 2.0 9:12 96 4.0 9:17 9:21 5 0
7 27 1.0 9:13 33 2.0 9:21 9:23 8 0
8 50 2.0 9:15 50 3.0 9:23 9:26 8 0
9 18 1.0 9:16 88 4.0 926 930 10
10 36 1.0 9:17 90 4.0 930 934
11 61 2.0 9:19 50 3.0 934 937
12 21 1.0 9:20 27 2.0 937 939
13 46 2.0 9:22 45 2.0 939 941
14 01 0 9:22 81 3.0 941 944
15 14 1.0 9:23 66 3.0 944 947

(c) Given the existing service level, does the drive-in window meet this criterion?

6
CB2203 Data-Driven Business Modeling

Question 3

The Three Hills Power Company provides electricity to a large metropolitan area
through a series of almost 200 hydroelectric generators. Management recognizes
that even a well-maintained generator will have periodic failures or breakdowns. As
expensive as the maintenance staff salaries are, breakdown expenses are even
higher. For each hour that one of its generators is down, Three Hills loses
approximately $75. Stephanie Robbins is the Three Hills Power Company
management analyst assigned to simulate maintenance costs. Robbins identifies two
important maintenance system components. First, the time between successive
generator breakdowns varies historically from as little as 0.5 hour to as much as 3
hours. For the past 100 breakdowns, she tabulates the frequency of various times
between machine failures:

Time Between Generator Breakdown at Three Hills Power

Time Between Recorded Number of Times Probabili
Machine Failures Observed ty
(Hours)
0.5 5 0.05
1.0 6 0.06
1.5 16 0.16
2.0 33 0.33
2.5 21 0.21
3.0 19 0.19
Total 100 1

The simulation of 15 generator breakdowns and the repair times required when one
repairperson was on duty per shift. The total simulated maintenance cost of the
previous system was $4,320. Robbins would now like to examine the relative cost-
effectiveness of adding one more repairing worker per shift. Each new repairperson
would be paid $30 per hour, the same rate as the first is paid. Robbins makes one
vital assumption as she begins—that repair times with two repairing workers will be
exactly one-half the times required with only one repairperson on duty per shift. In
the following table, she performs a statistical analysis of past repair times:

Repair time required with two-person

crews
Repair Time Required Probability
(Hours)
0.5 0.28
1 0.52
1.5 0.2
Total 1

7
CB2203 Data-Driven Business Modeling

List 4 – Random numbers to determine the time between breakdowns. Start

from the top of the list. There may be more random numbers than you
need.
69
84
12
94
51
36
17
02
15
29
16
52
56
43
26
22
08
62
40
67

List 5 – Random numbers to determine the required time of generator

repair. Start from the top of the list. There may be more random numbers
than you need.
37
77
13
10
02
18
31
19
32
85
31
94
81
43
31
58
33
51
29
11

8
 CB2203 Data-Driven Business Modeling

(a) Assume the first day begins at midnight (00:00 hours), fill in the following three
tables to design a SIMULATION BY HAND proposed maintenance system change
over a 15-generator breakdown period. Select the random numbers needed for
time between breakdowns from List 4. Select random numbers for generator
repair times from List 5.

Time Between Generator Breakdown at Three Hills Power

Time Between Number of Times Probabili Cumulativ Random
Recorded Machine Observed ty e Number
Failures (Hours) Probabilit Assignment
y
0.5 5 0.05
1.0 6 0.06
1.5 16 0.16
2.0 33 0.33
2.5 21 0.21
3.0 19 0.19
Total 100 1

Repair time required with two-person crews

Repair Time Probability Cumulative Random
Required (Frequency Probability Number
(Hours) ) Assignment
0.5 0.28
1 0.52
1.5 0.2
Total

Breakdo Rando Time Time of Time Rando Repair Time No.

wn m Between Breakdo Repair- m Time Repai Hours
Number Numbe Breakdown wn Person is Numbe Require r Machine
r (Hours) Free to r d Ends Down
Begin This (Hours)
Repair

9
CB2203 Data-Driven Business Modeling

(b) Calculate the cost of labor hours and cost of machine downtime. Measure the
simulated total maintenance cost.

(c) Should Three Hills add a second repairperson each shift?

10
CB2203 Data-Driven Business Modeling

Question 4

Dr. Mark Coleman is a dentist who owns his dental clinic. Coleman tries hard to
schedule appointments so that patients do not have to wait beyond their
appointment time. His October-20th schedule is shown in the following table.

Patient Scheduled Appointment Expected Time

Name Time Needed
Adams 9:30 a.m. 15
Brown 9:45 a.m. 20
Clark 10:15 a.m. 15
Dannon 10:30 a.m. 10
Edwards 10:45 a.m. 30
Foster 11:15 a.m. 15
Gray 11:30 a.m. 20
Harris 11:45 a.m. 15

Unfortunately, not every patient arrives exactly on schedule, and expected times to
examine patients are just that—expected. Some examinations take longer than
expected, and some take less time.

Greenberg’s experience dictates the following:

(a) 20% of the patients will be 20 minutes early.
(b) 10% of the patients will be 10 minutes early.
(c) 40% of the patients will be on time.
(d) 25% of the patients will be 10 minutes late.
(e) 5% of the patients will be 20 minutes late.

He further estimates that

(a) 15% of the time he will finish in 20% less time than expected.
(b) 50% of the time he will finish in the expected time.
(c) 25% of the time he will finish in 20% more time than expected.
(d) 10% of the time he will finish in 40% more time than expected.

List 6 – Random numbers to determine the patients’ arrival time. Start from
the top of the list. There may be more random numbers than you need.
60
08
19
29
36
72
30
27
50
64

11
 CB2203 Data-Driven Business Modeling

List 7 – Random numbers to determine the examination time. Start from the
top of the list. There may be more random numbers than you need.
80
45
86
99
02
34
87
08
86
84

(a) Assume that Dr. Coleman is ready to start his workday at 9:30 a.m. and that
patients are treated in order of their scheduled exam, fill in the following forms
to design a SIMULATION BY HAND.

Arrival Probabilit Cumulativ Random

Distribution y e Number
Probabilit Assignment
y

Examination Probabilit Cumulativ Random

Time y e Number
Distribution Probabilit Assignment
y

Patien Random Arrival Random Examination Time In Time

t Number Time Number Length (a.m.) Patient
(a.m.) (Minutes) Leaves

(b) Dr. Coleman has to leave at 12:10 p.m. on October 20 to catch a flight to a
dental convention in New York, otherwise he will miss his flight. Will he be able

12
CB2203 Data-Driven Business Modeling

to make the flight? Comment on this simulation and explain what factor
contributes to this situation? (Hint: did patient(s) arrive on time? Is the
estimated examination time accurate?)

13