IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING

Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 1: SHORTEST ROUTE PROBLEM

LEVEL: EASY

The Shortest Route Problem is a popular problem in Operations Research. It is an application of network models. The
problem consists of n nodes with a cost of travelling from one node to another. It is solvable via Djikstra’s Algorithm. It
may also be solved via trial-and-error.

Found below is a network model wherein the cost to travel from one node to another is given. A route must be created
from Node A to Node K.

4
C F
5
4 3

B I
2 6 6 2
2

A D G K
2
7
3 4
1 1
C J
8
9
2
E H
5

Question: What is the cheapest route from Node A to Node K?

(e.g. A-C-E-H-J-K; 20)
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 2: MINIMUM SPANNING TREE PROBLEM

LEVEL: EASY

The Minimum Spanning Tree Problem is a popular problem in Operations Research. It is an application of network
models. The problem consists of n nodes with a cost of travelling from one node to another. It is solvable via Prim’s
Algorithm. It may also be solved via trial-and-error.

Found below is a network model wherein the cost to travel from one node to another is given. A network that will
connect all nodes must be created.

N=4
U=4
S=1
O=5 T=3
H=9

J=9 P=9
R=10
F=2 M=7

A=6 C=4 K=8

E=2
Q=18
I=9
B=3

G=9 L=8

D=9

Question: What is the cheapest network that will connect all nodes?
(e.g. A, H, N, U, T, P, K, E, G; 54)
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 3: TRAVELLING SALESMAN PROBLEM

LEVEL: DIFFICULT

The Travelling Salesman Problem is a popular problem in Operations Research. It is an application of network models.
The problem consists of n cities with a cost matrix describing the cost of travelling from one city to another (and vice
versa). There is no algorithm that can efficiently solve the optimal answer to this problem. For small problem sizes, it
may be solved via trial-and-error.

In the province of Manira, there are seven cities. A salesman has to travel through each one of the seven cities and
return to his home city. He can start anywhere as long as he returns to where he started. He must incur the least cost
possible in doing this. The cost to travel from one city to another is found in the diagram and cost matrix below.

4
City G City F 2
8
5
City A
City E
9
6
2 9 12
7 4 10
5
8
City B
City D
4
City C 3

Summary of travelling cost:

FROM-TO A B C D E F G
A - 2 - 12 5 - 8
B 2 - 4 8 - 7 9
C - 4 - 3 4 6 5
D 12 8 3 - 10 9 -
E 5 - 4 10 - 2 -
F - 7 6 9 2 - 4
G 8 9 5 - - 4 -

Question: What is the cheapest route for the salesman to travel all seven cities?
(e.g. A-B-C-D-E-F-G-A; 33)
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 4: ASSIGNMENT PROBLEM

LEVEL: EASY

The Assignment Problem is a popular problem in Operations Research. It is an application of network models. The
problem consists of m workers and n assignments with a cost matrix describing the cost of assigning a worker to an
assignment. It is solvable via the Hungarian Algorithm. It may also be solved via trial-and-error.

Chootie Inc. is a tutorial center. For a particular day, six tutors are available: Armand, Bridgette, Cathy, Damien,
Edward, and Flavie. Also for that particular day, six tutees are asking for tutorial service: Ginny, Harry, Iman, John,
Kurt, and Lily. Chootie Inc. employs a customer-centric payment system. The tutees pay the company depending on
how happy they are with their tutor. The tutees rated the tutors from 1-10 (from previous sessions) depending on how
much they like the tutor, and will pay their tutors Php 100 times their rating i.e. if Ginny rated Armand with 6 and
Armand tutors Ginny, then Ginny pays the company Php 600. Below is the ratings matrix. The tutees should be
assigned one-to-one to each tutor.

Ratings Armand Bridgette Cathy Damien Edward Flavie

Matrix
Ginny 6 6 7 7 6 5
Harry 7 6 7 8 10 6
Iman 4 2 5 6 8 9
John 6 8 6 7 7 6
Kurt 5 4 2 5 7 5
Lily 6 3 4 6 5 3

Question: What is the best set of assignments that will yield the maximum profit for the company?
(e.g. Ginny-Armand, Harry-Bridgette, Iman-Cathy, John-Damien, Kurt-Edward, Lily-Flavie; Php 3400)
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 5: ART GALLERY PROBLEM

LEVEL: DIFFICULT

The Art Gallery Problem is a popular problem in Operations Research. The problem consists of an n-sided polygon to
be guarded by a minimum number of guards. There is no algorithm that can efficiently solve the optimal answer to this
problem. Non-optimal answers can be gotten using computational geometry. For small problem sizes, it may be
solved via trial-and-error.

The layout of an art gallery is found below. It shall be guarded by guards who shall be stationed at corners. The guard
can see anything with its 360o peripheral vision except when a wall is blocking.

QUESTION: What is the minimum number of guards needed to guard the whole art gallery?
(e.g. see figure on the right; 8)
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 6: 3-PARTITION PROBLEM

LEVEL: EASY

The 3-Partition Problem is a popular problem in Operations Research. The problem consists of n items with individual
values to be divided as equally as possible into 3 partitions. It can be solved using dynamic programming. It may also
be solved via trial-and-error.

A rich grandmother passes away. She owns the items below. The items must be shared between her three sons but
they must all get approximately the same value.

 Amethyst: Php 5M
 Carnelian: Php 35M
 Diamond: Php 3M
 Emerald: Php 6M
 Garnet: Php 15M
 Jade: Php 9M
 Opal: Php 12M
 Pearl: Php 13M
 Pyrite: Php 10M
 Quartz: Php 26M
 Ruby: Php 25M
 Sapphire: Php 20M
 Skull: Php 11M
 Topaz: Php 40M

QUESTION: What is the 3-way partition that minimizes the absolute difference between the values of the three parts?
(e.g. Partition 1: Amethyst, Carnelian, Diamond, Emerald, Garnet – Php 64M; Partition 2: Jade, Opal, Pearl, Pyrite,
Quartz – Php 70M; Partition 3: Ruby, Sapphire, Skull, Topaz – Php 96M; total 3-way difference: Php 64M)
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 7: KNAPSACK PROBLEM

LEVEL: EASY

The Knapsack Problem is a popular problem in Operations Research. The problem consists of items each with a weight
and a value. It can be solved using dynamic programming. It may also be solved via trial-and-error.

The sales manager for a publisher of a college textbook has 8 travelling salespeople to assign to four different regions
of the country. She has decided that each region should be assigned at least one salesperson. The table below gives
the estimated sales in each region if it were allocated a number of salespeople.

Number of Region
Salesperson 1 2 3 4
1 40 20 10 50
2 60 50 20 60
3 80 60 70 70
4 90 70 80 80
5 110 80 100 90

QUESTION: How many salespeople should be assigned to each region to maximize sales?
(e.g. Region 1: 5, Region 2: 1, Region 3: 1, Region 4: 1; 200)
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 8: TWO WORK CENTER SEQUENCING PROBLEM

LEVEL: DIFFICULT

The Two Work Center Sequencing Problem is a popular problem in Operations Research. It is a specific problem in
scheduling. The problem consists of n jobs each having a processing time in work center 1 and work center 2. There is
no algorithm that can efficiently solve the optimal answer to this problem. Non-optimal answers can be gotten using
Johnson’s Rule. For small problem sizes, it may be solved via trial-and-error.

Six jobs are to be done. The processing time of each job in work center 1 and work center 2 is given below. A work
center can only process one job at a time. Each job must undergo both processes in work center 1 and work center 2 in
that particular order (i.e. it must first be processed in work center 1 before in work center 2). The goal is to minimize
makespan i.e. the total time from when the first job starts being processed in work center 1 up to when the last job
ends being processed in work center 2.

Job Work Center 1 Work Center 2

A 5 5
B 4 3
C 8 9
D 2 7
E 6 8
F 12 15

Question: What should be the sequence of processing such that makespan is minimized?
(e.g. see illustration below: A-B-C-D-E-F; 33)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
WC1 A B C D E F
WC2 A B C D E F
 IE 3: INTRODUCTION TO INDUSTRIAL ENGINEERING
Activity 11: Problems Solving

Section:
Group Members:

PROBLEM 9: NEWSBOY PROBLEM

LEVEL: EASY

The Newsboy Problem is a popular problem in Operations Research. It is a specific problem in inventory management.
The problem consists of an item with a probabilistic demand. It can be solved using decision theory. It may also be
solved via trial-and-error.

Baker’s Fable is a very small bakery. In a day, it can sell 0 to 5 cakes with probabilities given below (i.e. 10% chance to
sell 0 cakes, 20% chance to sell 1 cake, etc). Baker’s Fable makes Php 300 profit per each cake sold during the day, but
it takes Php 100 loss per each cake not sold after the day (it expires after the day so the ingredients used were wasted).

Demand 0 1 2 3 4 5
Probability 10% 20% 30% 20% 10% 10%

Question: How many cakes should Baker’s Fable bake to maximize expected profit?
(e.g. see computation below: 5 cakes; Php 420)

If 5 bakes are caked: Cakes sold Profit from Cakes wasted Loss from Net profit for
Probability Demand cakes sold cakes wasted the day:
10% 0 0 Php 0 5 Php 500 -Php 500
20% 1 1 Php 300 4 Php 400 -Php 100
30% 2 2 Php 600 3 Php 300 +Php 300
20% 3 3 Php 900 2 Php 200 +Php 700
10% 4 4 Php 1200 1 Php 100 +Php 1100
10% 5 5 Php 1500 0 Php 0 +Php 1500
Expected payoff: Php 420 (sum all: probability x net profit)