MAT-485: Operations Research

Historical and Philosophical Summary of Operations Research (OR):

 Partly because the scientists involved were talented men, partly because of the pressures of
wartime necessity, and partly because of the synergism generated from the interactions of
different disciplines, these teams of scientists were remarkably successful in improving the
effectiveness of complex military operations.
 Examples of typical projects were radar deployment policies, antiaircraft fire control, fleet convoy
sizing, and detection of enemy submarines.
 After the war, many of the scientists who had been active in the military OR groups turned their
attention to the possibilities of applying a similar approach to civilian problems. By the early
1950s, these individuals had introduced the use of OR to a variety of organizations in business,
industry and government.
 It is interesting to note that the modern perception of OR as a body of established models and
techniques, that is, a discipline in itself- is quite different from the original concept of OR as an
activity, which was performed by interdisciplinary teams.

Operations Research (OR):

OR is an Experimental and applied science devoted to observe, understand, and predict the behavior of
purposeful man-machine systems. The OR workers are actively engaged in applying this knowledge to
practical problems in business, government and society.
 Mathematical modeling by example

Example 1:
Product mix: A toy company makes two types of toys: toy soldiers and trains. Each toy is produced in
two stages, first it is constructed in a carpentry shop, and then it is sent to a finishing shop, where it is
varnished, waxed, and polished. To make one toy soldier costs $10 for raw materials and $14 for labor;
it takes 1 hour in the carpentry shop, and 2 hours for finishing. To make one train costs $9 for raw
materials and $10 for labor; it takes 1 hour in the carpentry shop, and 1 hour for finishing.

There are 80 hours available each week in the carpentry shop, and 100 hours for finishing. Each toy
soldier is sold for $27 while each train for $21. Due to decreased demand for toy soldiers, the company
plans to make and sell at most 40 toy soldiers; the number of trains is not restricted in any way. What is
the optimum (best) product mix (i.e., what quantities of which products to make) that maximizes the
profit (assuming all toys produced will be sold)?
Here,
1. Decision Variables: 𝑥1 =number of toy soldiers and 𝑥2 =number of toy trains.

2. Variables Domains: The numbers 𝑥1 , 𝑥2 of toy soldiers and trains must be non-negative (sign
restriction), hence we get
0 ≤ 𝑥1 ≤ 40 and 𝑥2 ≥ 0.
3. Objective Function: The function that is to maximize/minimize:
𝑧 = (27 − 10 − 14)𝑥1 + (21 − 9 − 10)𝑥2 , that is, the profit for selling 𝑥1 toy soldiers and 𝑥2 toy
trains is
𝑧 = 3𝑥1 + 2𝑥2

4. Goal/Objective: Maximize the profit, that is, maximize the function 𝑧 = 3𝑥1 + 2𝑥2

5. Constraints: Producing 𝑥1 toy soldiers and 𝑥2 toy trains require

(a) 1𝑥1 + 1𝑥2 hours in the carpentry shop; there are 80 hours available
(b) 2𝑥1 + 1𝑥2 hours in the finishing shop; there are 100 hours available

Therefore, the problem is

Maximize the profit function 𝑧 = 3𝑥1 + 2𝑥2

where
𝑥1 + 𝑥2 ≤ 80
2𝑥1 + 𝑥2 ≤ 100
𝑥1 ≤ 40
𝑥1 , 𝑥2 ≥ 0

We call this a program. It is a linear program, because the objective is a linear function of the decision
variables, and the constraints are linear inequalities (in the decision variables).
MATHEMATICAL PROGRAMMING

 Programming problems in general are concerned with the use or allocation of scarce resources-
labor, materials, machines, and capital- in the “best” possible manner so that costs are
minimized or profits are maximized.
 Often these situations are mathematical representations of real-world problems, and hence the
term mathematical programming.

LINEAR PROGRAMMING

The term linear programming merely defines a particular class of mathematical programming problems
that meet the following conditions:

 The decision variables involved in the problem are nonnegative.

 The criterion for selecting the “best” values of the decision variables can be described by a linear
function of these variables, that is, a mathematical function involving only the first powers of
the variables with no cross products. The criterion function is normally referred to as the
objective function.
 The operating rules governing the process can be expressed as a set of linear equations or linear
inequalities. This set is referred to as the constraint set.

Motivation: Managements trying to make the most effective use of an organization’s resources.

Mathematical programming is selecting the best option(s) from a set of alternatives mathematically by
minimizing/maximizing a function.

Resources: machinery, labor, money, time, warehouse space, or raw materials.

Resources may be used

(i) To produce products: machinery, furniture, food, or clothing Or

(ii) To provide services: schedules for shipping and production, advertising policies, or investment
decisions.
GENERAL FORM OF A LINEAR PROGRAMMING PROBLEM

n
( LP) Max / Min z   ci xi
i 1
n
subject to  a x {, , }b
j 1
ij j i (i  1,..., m)

Introduction: Problems in OR: In the world of OR, programming refers to modeling and solving a
problem mathematically.

 Linear Programming (LP)

 Nonlinear Programming (NLP)

 Integer Programming (IP)

 Allocation

 Replacement

 Routing

 Game

 Transportation

 Inventory

 Sequencing and Many More

Linear Programming (LP)

 LP is a widely used mathematical technique designed to help managers in planning and decision

making relative to resource allocation.

 Formulation
  Graphical Method

 Simplex Method

 Dual Simplex Method

 Duality of LP

 Sensitivity Analysis

 Transportation Problems

 Assignment Problems

Brief History of LP

• 1st initiative was taken by Leonid Kantorovich, 1939 during the World War ll.
• Modern day LP started with the work of George Dantzig, 1947.
• Dantzig developed the simplex algorithm in 1947
• Over the last 30 years, LP became popular
• John Von Newman developed the theory of duality.
• LP has been in vast use in real life problems

Recall: Similar to Example 1 Problem

A company wishes to schedule the product of a kitchen appliance that requires two resources: labor and

raw materials. The company is considering three different types of products (A, B, C). To produce one unit

of product A, it requires 7 hours of labor and 4 pounds of raw materials. To produce one unit of product

B, it requires 3 hours of labor and 4 pounds of raw materials. To produce one unit of product C, it requires

6 hours of labor and 5 pounds of raw materials. Profit from each unit of product A is Tk. 4. Profit from

each unit of product B is Tk. 2. Profit from each unit of product C is Tk. 3. The supply of raw materials is

restricted to 200 pounds per day. The daily availability of labor is 150 hours.

Formulate an LP model to determine the daily production rate of the various types of products in order

to maximize the total profit.

The formulated LP is as follows:

Mazimize 𝑍 = 4𝑥1 + 2𝑥2 + 3𝑥3

Subject to the constraints 7𝑥1 + 3𝑥2 + 6𝑥3 ≤ 150

4𝑥1 + 4𝑥2 + 5𝑥3 ≤ 200

𝑥1 , 𝑥2 , 𝑥3 ≥ 0

Art of Formulation: Building LP models is one of the most creative aspects of OR.

To formulate LP models correctly we should

 read the problem statement very carefully

 understand the problem
 Identify the decision variables with units of measure
 Pulled out numerical values, present them in a table or set of tables
 Form objective function and the constraints

Three basic steps in formulation:

1. Identify the unknown variables

2. Identify all the restrictions or constraints
3. Identify the objective

Linear Program in Standard Form:

The standard form of a linear programming problem with 𝑚 equations and

𝑛 variables can be represented as follows:

Maximize (Minimize): 𝑍 = 𝑐1 𝑥1 + 𝑐2 𝑥2 + ⋯ + 𝑐𝑛 𝑥𝑛

Subject to : 𝑎11 𝑥1 + 𝑎12 𝑥2 + ⋯ + 𝑎1𝑛 𝑥𝑛 = 𝑏1

𝑎21 𝑥1 + 𝑎22 𝑥2 + ⋯ + 𝑎2𝑛 𝑥𝑛 = 𝑏2

⋮
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + ⋯ + 𝑎𝑚𝑛 𝑥𝑛 = 𝑏𝑚
𝑥1 ≥ 0, 𝑥2 ≥ 0, … , 𝑥𝑛 ≥ 0

𝑏1 ≥ 0, 𝑏2 ≥ 0, … , 𝑏𝑚 ≥ 0
Main Features of the Standard Form:

(a) The objective function is of the maximization or minimization type.

(b) All constraints are expressed as equations, except the non-negative restrictions.

(c) All variables are restricted to be non-negative.

(d) The right-hand side constant of each constraint is non-negative.

(e) If 𝑥𝑖 ≤ 0, then put 𝑥𝑖 = −𝑥𝑖′ . Clearly, 𝑥𝑖′ ≥ 0.

(f) If 𝑥𝑖 is unrestricted in sign (i.e. a free variable), then put 𝑥𝑖 = 𝑥𝑖′ − 𝑥𝑖′′ , where 𝑥𝑖′ ≥ 0, 𝑥𝑖′′ ≥ 0.

Matrix-Vector Notation:

In matrix-vector notation, the standard linear programming problem can be expressed

in a compact form as

Maximize (Minimize) : 𝑍 = 𝒄𝒙

Subject to : 𝐴𝒙 = 𝒃

𝒙≥𝟎

𝒃≥𝟎
where 𝐴 is an 𝑚 × 𝑛 matrix, 𝒙 is an (𝑛 × 1) column vector, 𝒃 is an (𝑚 × 1) column
vector, and 𝒄 is a (1 × 𝑛) row vector.

In other words,
𝑥1 𝑏1
𝑥2 𝑏2
𝑎11 𝑎12 … 𝑎1𝑛 . .
𝑎21 𝑎22 … 𝑎2𝑛 𝒙𝑛×1 = . 𝒃𝑚×1 =
.
𝐴𝑚×𝑛 =[ … … … … ] . .
𝑎𝑚1 𝑎𝑚2 … 𝑎𝑚𝑛 [𝑥𝑛 ] [𝑏𝑚 ] 𝒄1×𝑛 = (𝑐1 , 𝑐2 , … , 𝑐𝑛 )
Coefficient Matrix Decision Vector Requirement vector Profit (Cost) vector
LP in Canonical Form:

Matrix-Vector Notation of Canonical Form:

Maximizing Problem:
In matrix-vector notation, the linear programming problem in general form
can be expressed in a compact form as

Maximize: 𝑍 = 𝒄𝒙

Subject to : 𝐴𝒙 ≤ 𝒃

𝒙≥𝟎
𝒃≥𝟎
where 𝐴 is an 𝑚 × 𝑛 matrix, 𝒙 is an (𝑛 × 1) column vector, 𝒃 is an (𝑚 × 1)
column vector, and 𝒄 is a (1 × 𝑛) row vector.
In summation form:

Minimizing Problem:

In matrix-vector notation, the linear programming problem in general form

can be expressed in a compact form as

Minimize: 𝑍 = 𝒄𝒙

Subject to : 𝐴𝒙 ≥ 𝒃

𝒙≥𝟎
𝒃≥𝟎

where 𝐴 is an 𝑚 × 𝑛 matrix, 𝒙 is an (𝑛 × 1) column vector, 𝒃 is an (𝑚 × 1)

column vector, and 𝒄 is a (1 × 𝑛) row vector.

Questions that should be answered:

1. What are the decision variables?

2. What are parameters/ inputs?

3. What is our objective?

4. How the objective is related to the decision variables?

5. Is it a maximization or minimization problem?

6. What are the constraints?

7. What requirements must be met?

8. Should we use inequality or equality type of constraints?

9. Are the variables integer?

Model Components:

 Decision variables (𝑥𝑖𝑗 )- mathematical symbols representing levels of activity of a firm.

 Objective function - a linear mathematical relationship describing an objective of the firm, in

terms of decision variables - this function is to be maximized or minimized.

 Constraints – requirements or restrictions placed on the firm by the operating environment,

stated in linear relationships of the decision variables.

 Parameters - numerical coefficients and constants used in the objective function and
constraints.

Definitions that we should know:

Exercise: Transform the following linear programming programs to the standard form:

Maximize 𝑍 = 2𝑥1 − 𝑥2 + 4𝑥3

Subject to −3𝑥1 + 𝑥2 = −4
𝑥1 − 2𝑥2 + 𝑥3 ≥ −3

2𝑥1 + 3𝑥3 ≤ 2
𝑥1 ≥ 0, 𝑥2 ≤ 0, 𝑥3 𝑢𝑛𝑟𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑒𝑑 𝑖𝑛 𝑠𝑖𝑔𝑛

Minimize 𝑍 = 2𝑥1 − 3𝑥2 + 𝑥3 − 4𝑥4

Subject to 𝑥1 − 𝑥2 + 𝑥3 − 2𝑥4 = −5

−2𝑥1 + 𝑥2 − 2𝑥3 + 3𝑥4 ≥ 3

3𝑥1 − 2𝑥2 + 𝑥3 − 5𝑥4 ≤ 4
𝑥1 ≤ 0, 𝑥2 ≥ 0, 𝑥3 ≥ 0,

𝑥4 𝑢𝑛𝑟𝑒𝑠𝑡𝑟𝑖𝑐𝑡𝑒𝑑 𝑖𝑛 𝑠𝑖𝑔𝑛
Product mix
A toy company makes two types of toys: toy soldiers and trains. Each toy is produced in two stages, first
it is constructed in a carpentry shop, and then it is sent to a finishing shop, where it is varnished, waxed,
and polished. To make one toy soldier costs $10 for raw materials and $14 for labor; it takes 1 hour in the
carpentry shop, and 2 hours for finishing. To make one train costs $9 for raw materials and $10 for labor;
it takes 1 hour in the carpentry shop, and 1 hour for finishing.
There are 80 hours available each week in the carpentry shop, and 100 hours for finishing. Each toy soldier
is sold for $27 while each train for $21. Due to decreased demand for toy soldiers, the company plans to
make and sell at most 40 toy soldiers; the number of trains is not restricted in any way. What is the
optimum (best) product mix (i.e., what quantities of which products to make) that maximizes the profit
(assuming all toys produced will be sold)?

The LP problem is
Maximize the profit function 𝑧 = 3𝑥1 + 2𝑥2
where
𝑥1 + 𝑥2 ≤ 80
2𝑥1 + 𝑥2 ≤ 100
𝑥1 ≤ 40
𝑥1 , 𝑥2 ≥ 0