Mathematics for Business

1Part 4: Mathematical Programming

Mathematical Programming

• Linear Programming
• Transportation Problem
• Decision Making
Mathematical Programming
1Linear Programming
Linear Programming
Linear Programming

Assumptions:
• Linearity
• Certainty
• Devisability
Examples
The xyz Company wishes to schedule the production of two items namely
chairs and tables. The management of the company has to decide as to how
many chairs and tables to be produced per day to maximize the profit. The
following information is available to the management. The profit per item is
$35 and $55, respectively. In order to produce the two items that requires
resources wood, manpower and machine hour. The supply of wood is
restricted to 1500 kg/day. The manpower available per day is 75 carpenters.
Also the maximum of 120 machine hours available per day. The production
formula is as given in the following table.

Formulate an LP model to determine the daily production rate of various

models in order to maximize the profit.
Examples
Examples
A firm manufactures three products A, B and C. The profits are $3, $2 and
$4, respectively. The firm has two machines. The processing time in minutes
for each machine on each product is given below.

Machines D and E have 2000 and 2500 machine minutes, respectively. The firm
must manufacture at least 100 A’s; 200 B’s and 50 C’s but not more than 150 C’s.
Formulate a linear programming model to maximize the total profit.
Examples
Formulate the following problem as an LP model:
A co-operative farm owns 100 acres of land and has $25,000 in funds available for
investment. The farm members can produce a total of 3500 man hours worth of labor
during the months September-May and 4000 man hours during June-August. If any
of these man-hours are not needed, some members of the farm will use them to work
on a neighbouring farm for $2/hour during September-May and $3/hour during
June-August. Cash income can be obtained from the three main crops and two types
of livestock: dairy cows and laying hens. No investment funds are needed for the
crops. However, each cow will require an investment outlay of $3200 and each hen
will require $15.
Moreover, each cow will require 1.5 acres of land, 100 man-hours of work during
September-May and another 50 man-hours during June-August. Each cow will
produce a net annual cash income of $3500 for the farm. The corresponding figures
for each hen are: no acreages, 0.6 man-hours during September-May, 0.4 man-hours
during June-August, and an annual net cash income of $200. The chicken house can
accommodate a maximum of 4000 hens and the size of the cattle-shed limits the
members to a maximum of 32 cows.
Examples
Estimated man-hours and income per acre planted in each of the three crops are:

The co-operative farm wishes to determine how much acreage should be planted in
each of the crops and how many cows and hens should be kept to maximize its net
cash income.
Examples
Relax and Enjoy Lake Development Corporation is developing a lake side
community at a privately owned lake and in the business of selling property for
vacation and rent cottages. The primary market for these lakeside lots are middle and
upper income families within 100 miles of the development. Relax and Enjoy has
employed the advertising firm Philippines Jackson to design the promotional
campaign for the project.
After considering possible advertising media and the market to be covered, the
agency has made the preliminary recommendations to restrict the first month’s
advertising to five sources. At the end of this month, the agency will re-evaluate the
strategy based upon the month’s results. The agency has collected data on the
number of potential purchase families reached, the cost per advertisement, the
maximum number of times each medium is available and the expected exposure for
each of the five media. The expected exposure is measured in terms of an exposure
unit, a management judgement measure of the relative value of one advertisement in
each of the media. The measures based on agency’s experience in the advertising
business take into account such factors as audience profile (age, income and
education of the audience reached), image presented and quality of the
advertisement. The information collected is presented in Table.
Examples
Advertising media alternative for relax and enjoy lake development corporation:

Relax and Enjoy has provided with an advertising budget of $300,000 for the first
month’s campaign.
In addition, the company has a wish to have at least 10 television commercials, and
atleast 50,000 potential purchasers must be reached during the month. Formulate the
problem of media selection.
Standard Form

If an LPP satisfies the following conditions, then it is said to be in a standard

form.
(i) All the variables are non-negative.
(ii) All constraints must be equations together with non-negative right-hand
side values.
(iii) The objective function may be either minimization or maximization.
Example
Find the standard form of an LPP:
Minimize Z = 3x1 + 2x2
s.t. 2x1 + x2 = 15; 3x1 – 4x2 ≥ –6; x1 – 2x2 ≥ 5; x2 ≥ 0
Sol.
Change: x1= x3 – x4, where x3, x4 ≥ 0

Change all the constraints ≥ to ≤ by multiple with –1:

–3(x3 – x4) – 4x2 ≤ 6
Change all the constraints ≤ to = by adding new variables
–3(x3 – x4) – 4x2 + x5 = 6 và x3 – x4 – 2x2 + x6 = 5
Methods for the LPP
Solutions for the LPP
Solutions
Basic Solution Assume that the standard form of an LPP has ‘m’ equations with ‘n’
variables (m ≤ n) together with Non-negative conditions. The unique solutions resulting
from setting (n - m) variables to zero such that the remaining m columns linearly
independent are called basic solutions.
Basic Feasible Solution If a basic solution satisfies the Non-negative conditions, it is
named as basic feasible solution.
Basic Infeasible Solution If a basic solution does not satisfy the Non-negative
conditions, it is named as basic infeasible solution.
Optimum Solution If a basic feasible solution maximizes or minimizes the objective
function of the problem, it is named as an optimum solution.
Unique Optimum Solution It means that the optimum value is attained for only one
feasible solution.
Multiple Optimum Solution It means that the same optimum value is attained for more
than one feasible solution.
Basic and Non-basic Variables The variables set equal to zero are (refer to basic
solution definition) called non-basic variables, the remaining variables are called basic
variables.
Degenerate Solution If one or more basic variables become zero, that solution is
named as degenerate solution.
Examples
Find all the basic feasible solutions to the system of linear
equations, state which among them are, (i) Basic feasible (ii)
Basic infeasible
1) x1 + 2x2 + x3 = 4; 2x1 + x2 + 5x3 = 5
2) 3x1 + x2 + x3 = 5; –x1 + x2 – x4 = 1
Convex Set
A set X is convex if for any two points x1, x2 in the set, the line segment joining
x1, x2 is also in the set, i.e, if X is convex, then x1, x2 X => lx1 + (1 - l)x2 X; 0 ≤ l
≤ 1.

A point x is an extreme point of a convex set iff there do not exist other points
x1, x2, x1 ≠ x2 in the set such that x = lx2 + (1 - l)x1, 0 < l < 1.
Eg.
1) In a triangle, the three vertices are extreme points.
2) In a circle, all boundary points are extreme points.
Remark 1: All boundary points need not be extreme points.
Remark 2: The maximum number of extreme points of a convex set of feasible
solution to a LPP is nCm, where ‘n’ is the number of variables and m be the number of
constraints in the problem. Such that m ≤ n.
Graphical Solutions
Examples
1) Max Z = 3x1 + 9x2; st. x1 + x2 ≤ 8; x1 + 2x2 ≤ 4, x1, x2 ≥ 0
2) Min Z = 2x1 + 3x2; s.t. x1 + x2 ≤ 4; 6x1 + 2x2 ≥ 8; x1 + 5x2 ≥ 4; x1 ≤ 3; x2 ≤ 3; x1, x2 ≥ 0
3) Max Z = 4x1 + 5x2 s.t. x1 + x2 ≥ 1; –2x1 + x2 ≤ 1; 4x1 – 2x2 ≤ 1; x1, x2 ≥ 0
4) Max Z = 6x1 − 2x2 st. 2x1 − x2 ≤ 0; x1 ≤ 4; x1, x2 ≥ 0
5) Max Z = x1 + x2 st. x1 + x2 ≤ 1; −3x1 + x2 ≥ 3; x1, x2 ≥ 0
6) Max Z = −0.1x1 + 0.5x2 s.t. x1 + x2 ≤ 20; 2x1 + 5x2 ≤ 80; −0.1x1 + 0.5x2 = 8; x1, x2 ≥ 0
7) Max Z = 5x+6y st. 4x+2y ≤ 10; 4x+2y ≥ 20; x, y ≥ 0
8) Min. Z = 4x+6y st. 4x+2y = 10; 4x+2y ≥ 20; x, y ≥ 0
9) Min Z = 5x + 7y st. 2x + 3y ≥ 6; 3x − y ≤ 15; −x + y ≤ 4; 2x + 5y ≤ 27; x,y ≥ 0
10) Min Z = 5x+5y, st. 2x+y ≥ 15; x+4y ≥ 25; x+2y ≥ 20; x, y ≥ 0
Given x,y ≥ 0, find Max and Min of:
11) Z = 2x + 3y st. 2x + 3y ≥ 6; 3x−y ≤ 15; −x + y ≤ 4; 2x + 5y ≤ 27
12) Z = 6x + 4y st. −x + y ≤ 12; x + y ≤ 24; 2x + 5y ≤ 80
13) Z = 6x + 5y st. x + y ≥ 6; 2x +y ≥ 9
14) Z = x – y st. x + y ≥ 6; 2x +y ≥ 9
Examples
1) A sick patient has been advised by his doctor that his diet must contain at least
4000 units of vitamins, 50 units of minerals and 1400 units of calories. Two foods,
F1 and F2, are available at a cost of $4 and $3 per unit, respectively. One unit of
food F1 contains 200 units of vitamins, 1 unit of mineral and 40 units of calories.
One unit of food F2 contains 100 units of vitamins, 2 units of minerals and 40 units
of calories. Formulate the above problem as an LPP and solve using graphical
method.
2) Memory Chip Data for CPUs:

How many of two types of high-speed computer chips should to be manufactured

weekly to maximize their profits?
Simplex Methods
(i) Simplex method-I or ordinary simplex method
(ii) Big-M method or penalty method or Charnes penalty method
(iii) Two-phase simplex method
(iv) Dual simplex method (DSM)
Ordinary Simplex Method
Notations Used
• B — Basis elements
• CB — Cost coefficient of the basic element in Cj row
• Cj — Coefficient of the corresponding variable in the objective function. (j = 1, 2,
…, n)
• xj — Variable (for j = 1, 2, ..., n)
• XB — Solution to the basic elements.
• Zj — CB xj ( j = 1, 2, ...,n) (Multiply the corresponding entries and add it)
• Zj – Cj — CB xj - Cj ( j = 1, 2, ..., n) (called net evaluation)
• PC — Pivotal column
• PR — Pivotal row
• PE — Pivotal element
• q — XB/PC
The Structure of a Simplex Table

Examples: Solve the following LPPs by simplex method

1) Max Z = 2x1 + 4x2; s.t. 2x1 + 3x2 ≤ 48; x1 + 3x2 ≤ 42; x1 + x2 ≤ 21; x1, x2 ≥ 0
2) Max Z = 4x1+10x2; s.t. 2x1 + x2 ≤ 50; 2x1 + 5x2 ≤ 100; 2x1 + 3x2 ≤ 90; x1, x2 ≥ 0
3) Max Z = 4x1 + x2 + 3x3 + 5x4 s.t. 4x1 – 6x2 – 5x3 + 4x4 ≥ - 20;
3x1 – 2x2 + 4x3 + x4 ≤ 10; 8x1 – 3x2 + 3x3 + 2x4 ≤ 20; x1, x2, x3, x4 ≥ 0
4) Max Z = 30x1 + 25x2, st. 25x1 + 20x2 ≤ 80; 20x1 + 25x2 ≤ 90; x1, x2 ≥ 0
5) Max Z = 10x1 + 5x2, st. 5x + 2x2 ≤ 34; 4x1 + 4x2 ≤ 40; x1, x2 ≥ 0
6) Max Z = 30x1 +25x2 + 20x3, st. 25x1 + 10x2 +15x3 ≤ 80; 20x1 + 25x2 + 10x3 ≤ 90; x1,
x2 , x 3 ≥ 0
7) Max Z = 3x1 + 2x2 + x3, st. 2x1 + x2 ≤ 60; 2x2 + x3 ≤ 80; 10x1 + 15x2 + 20x3 ≤ 100;
x + x + 2x ≤ 90; x , x , x ≥ 0.
 Examples
1) Max Z = 400x1 + 100x2 s.t. 4x1 + 2x2 ≤ 1600; 5/2x1 + x2 ≤ 1200; 9/2x1 + 3/2x2 ≤ 1600; x1, x2 ≥
0
2) Max Z = 12x1 + 3x2 + x3 s.t. 10x1 + 2x2 + x3 ≤ 100; 7x1 + 3x2 + 2x3 ≤ 77; 2x1 + 4x2 + x3 ≤ 80;
x 1, x 2 , x 3 ≥ 0
3) Max Z = 20x1 + 6x2 + 8x3 s.t. 8x1 + 2x2 + 3x3 ≤ 250; 4x1 + 3x2 ≤ 150; 2x1 + 3x3 ≤ 90; x1, x2, x3
≥0
4) Max Z = 50x1 + 20x2 + 30x3 s.t. 5x1 + x2 + 7x3 ≤ 5; 5x1 + x2 + 6x3 ≤ 6; 3x1 - x2 - 9x3 ≤ 3; x1,
x 2, x 3 ≥ 0
5) Max Z = 2x1 + 5x2 s.t. x1 + 4x2 ≤ 24; 3x1 + x2 ≤ 21; x1 + x2 ≤ 9; x1, x2 ≥ 0
6) Max Z = 4x1 + 5x2 + 9x3 + 11x4 s.t. x1 + x2 + x3 + x4 ≤ 15; 7x1 + 5x2 + 3x3 + 2x4 ≤ 120; 3x1 +
5x2 + 10x3 + 15x4 ≤ 100; x1, x2, x3, x4 ≥ 0
7) Max Z = 3x1 + 4x2 + x3 + 7x4 s.t. 8x1 + 3x2 + 4x3 + x4 ≤ 7; 2x1 + 6x2 + x3 + 5x4 ≤ 3; x1 + 4x2 +
5x3 + 2x4 ≤ 8; xj ≥ 0 (j = 1, ..., 4)
8) Max Z = 4x1 + 5x2 s.t. x1 + 3x2 ≤ 12; 4x1+ 3x2 ≤ 24; x1, x2 ≥ 0;
9) Max Z = x1 + 2x2 s.t. -x1 + 2x2 ≤ 8; x1 + 2x2 ≤ 12; x1 - 2x2 ≤ 3; x1, x2 ≥ 0;
10) Max Z = 50x1 + 70x2 s.t. 120x1 + 120x2 ≤ 8400; x1 + 2x2 ≤ 100; 2x1 + x2 ≤ 120; x1, x2 ≥ 0
 Examples
1) Suppose that a small pharmaceutical firm produces three types of drugs, x1, x2, and x3, using three types of
chemicals, A, B, and C, the maximum available quantities of which are 45, 40, and 25 units, respectively. The
firm needs to use 1, 0, and 2 units, respectively, of A, B, and C to produce one unit of x1; 0, 2, and 1 units of A,
B, and C, respectively, to produce one unit of x2; and 3, 2, and 0 units of A, B, and C, respectively, to produce
one unit of x3. The revenue that the firm obtains from one unit of x1, x2, and x3 is $6, $5, and $3, respectively.
Find the quantities of the drugs that the firm must produce to maximize its revenue.
2) Assume that a company employs three salespeople, A, B, and C, temporarily to sell its three products, x1, x2,
and x3. It was directed that 1, 0, and 2 hours of A, B, and C to be spent for the sale of 1 unit of x1; 0, 2, and 1
hours of A, B, and C to be spent for the sale of 1 unit of x2; and 4, 2, and 0 hours of A, B, and C to be spent for
the sale of 1 unit of x3. The minimum number of hours to be spent by A, B, and C must be 40, 45, and 30 hours,
respectively. The cost of sale of one unit of x1, x2, and x3 is $4, $3, and $2, respectively. How many units of the
three goods should be sold so that the company’s total cost will be minimum?
3) An auto company manufactures cars and trucks. Each vehicle must be processed in the paint shop and body
assembly shop. If the paint shop where only painting truck, takes place 40 per day could be painted. If the paint
shop where only painting car, 60 per day could be painted. If the body shop where only producing cars, it could
process 50 per day. If the body shop where only producing trucks, it could process 50 per day. Each truck
contributes $10,000 to profit, and each car contributes $8000 to profit. Use linear programming to determine a
daily production schedule that will maximize the company’s profits.
Big-M Method
Examples
1) Min Z = 12x1 + 20x2; s.t. 6x1 + 8x2 ≥ 100; 7x1 + 12x2 ≥ 120; x1, x2 ≥ 0
2) Max Z = 3x1 + 2x2; s.t. 2x1 + x2 ≤ 2; 3x1 + 4x2 ≥ 12; x1, x2 ≥ 0
3) Max Z = 8x2; s.t. x1 – x2 ≥ 0; 2x1 + 3x2 ≤ 6; x1, x2 unrestricted.
4) Max Z = 6x1 + 4x2; s.t. 2x1 + 3x2 ≤ 30; 3x1 + 2x2 ≤ 24; x1 + x2 ≥ 3; x1, x2 ≥ 0
5) Min C = 8x1 +10x2, st. x1 +3x2 ≥ 30; 4x1 +x2 ≥ 20; x1, x2 ≥ 0;
6) Min C = 5x1 +3x2, st. 6x1 + 3x2 ≥ 30; x1 + 4x2 ≥ 20; x1, x2 ≥ 0;
6) Min C = 4x1 + 3x2 + 2x3, st. x1 + 4x2 + x3 ≥ 40; 5x1 + 4x2 +x3 ≥ 50; 2x1 + x2 + x3 ≥ 60;
x1, x2, x3 ≥ 0;
7) Min C = 8x1 + 5x2 + 3x3, st. 4x2 + x3 ≥ 20; 5x1 + 4x3 ≥ 40; 4x1 + 4x2 ≥ 40; x1, x2, x3 ≥
0
8) Max Z = 2x1+ 3x2 s.t. x1 + 2x2 ≤ 4; x1 + x2 = 3; x1, x2 ≥ 0
9) Min Z = 5x + 6y; s.t. 2x + 5y ≥ 1500; 3x + y ≥ 1200; x, y ≥ 0
10) Max Z = 2x1 + x2 + x3 s.t. x1 + x2 + 2x3 ≤ 5; 2x1 + 3x2 + 4x3 = 12; x1, x2, x3 ≥ 0
11) Max Z = 2x1 + x2; s.t. x1 - x2 ≤ 10; x1 ≤ 20; x1, x2 ≥ 0
12) Max Z = x1 + 2x2 + 3x3 - x4 s.t. x1 + 2x2 + x3 = 15; 2x1 + x2 + 5x3 = 20; x1 + 2x2 + x3
+ x = 10; x , x , x , x ≥ 0
 Examples
1) Three grades of coal A, B and C contains phosphorus and ash as impurities. In a
particular Industrial process, fuel upto 100 ton (Maximum) is required which should
contain ash not more than 3% and phosphorus not more than 0.03%. It is desired to
maximize the profit while satisfying these conditions. There is an unlimited supply of
each grade. The percentage of impurities and the profits of grades are given below:

Find the proportion in which the three grades be used.

2) Assume that a company employs three salespeople, A, B, and C, temporarily to sell
its three products, x1, x2, and x3. It was directed that 1, 0, and 2 hours of A, B, and C to
be spent for the sale of 1 unit of x1; 0, 2, and 1 hours of A, B, and C to be spent for the
sale of 1 unit of x2; and 4, 2, and 0 hours of A, B, and C to be spent for the sale of 1 unit
of x3. The minimum number of hours to be spent by A, B, and C must be 40, 45, and 30
hours, respectively. The cost of sale of one unit of x1, x2, and x3 is $4, $3, and $2,
respectively. How many units of the three goods should be sold so that the company’s
total cost will be minimum?
Two Phases Method
 Eg. Solve the following problem using two-
phase simplex method:

1) Min. Z = x1+ x2; s.t. 2x1 + 4x2 ≥ 4; x1 + 7x2 ≥ 7; x1, x2 ≥ 0

2) Min Z = -x1 - 2x2 - 3x3 subject to -2x1 + x2 + 3x3 = 2; 2x1 + 3x2 + 4x3 = 1; x1, x2, x3 ≥ 0
3) Max Z = 2x1 + x2 + x3 s.t. 4x1 + 6x2 + 3x3 ≤ 8; 3x1 - 6x2 - 4x3 ≤ 1; 2x1 + 3x2 - 5x3 ≥ 4;
x 1, x 2 , x 3 ≥ 0
4) Max Z = 5x1 - 4x2 + 3x3 s.t. 2x1 + x2 - 6x3 = 20; 6x1 + 5x2 + 10x3 ≤ 76; 8x1 - 3x2 + 6x3 ≤
50; x1, x2, x3 ≥ 0
5) Min Z = 2x1 + x2 + 3x3 s.t. 5x1 - 2x2 ≥ -5; 3x1 + 2x2 + 4x3 ≥ 7; 2x1 + 5x3 ≤ 3; x1, x2, x3 ≥ 0
6) Max Z = 5x1 - x2 s.t. 2x1 + x2 = 6; x1 + x2 ≤ 4; x1 + 2x2 ≤ 5; x1, x2 ≥ 0
7) Min Z = -3x1 + x2 s.t. x1 - 2x2 ≥ 2; -x1 + x2 ≥ 3; x1, x2 ≥ 0;
Example

1) M/s Funschool Toys Ltd. manufactures two models of toys T1 and T2. Each of
these products must go through two manufacturing processes A and B. Each unit of T1
requires one hour in Process A and 1. hours in Process B; each unit of T2 requires 2
hours in Process A and 3/4 hour in process B. The machinery supporting Process A
currently has a weekly machine hour capacity of 100 hours, and the weekly capacity
of process B machinery is 90 machine hours. Per unit profits on T1 and T2 are $100
and $150 respectively. Determine the optimal product mix for Funschool Ltd.
2) The emergency Ward of M/s. Apollo Hospitals Ltd., schedules Nurses to come on
duty every 3 hours for a 6 hours shift. The shift times and the minimum number of
nurses required during each 3 hours time interval are given below. Determine the
schedule that will minimize the total number of nurses employed in said Emergency
ward.
Dual problems

• The dual problem:

Dual problems

• The dual problem:

Relationships between primal and dual LP
problems
• If the primal LP problem is a maximization (minimization) problem, then its dual
is a minimization (maximization) problem. This means that we have to change the
≤ (≥) sign to ≥ (≤) in the constraints when we set up the dual of a primal
maximization (minimization) problem.
• The matrix of the coefficients of the constraints of the dual problem is the
transpose of the matrix of the coefficients of the constraints of the primal problem.
Therefore, when we set up the dual of a LP problem we need to transpose the
matrix of the coefficients of the constraints.
• The column vector of the constants (the RHS) in the dual problem is the transpose
of the row vector of the coefficients of the objective function in the primal
problem. This implies that when we set up the dual of a LP problem we need to
transpose the coefficients of the objective function in the primal problem to form
the column vector of the constants in the dual problem.
• The row vector of the coefficients of the objective function in the dual problem is
the transpose of the column vector of constants (the RHS) in the primal problem.
Therefore, when we set up the dual of a LP problem we need to transpose the
column vector of constants in the primal problem to obtain the row vector of the
coefficients of the objective function in the dual problem.
• The dual of a dual LP problem is the primal LP problem.
• The variables in both the primal problem and the dual problem are nonnegative.
Relationships between primal and dual LP
problems
 Examples: Find the dual problem

1) Max. z = 80x1 + 90x2 s.t. 3x1 + 5x2 ≤ 30; 3x1 + 2x2 ≤ 21; x1, x2 ≥ 0
2) Min. z = 10x1 + 8x2 s.t. x1 + 2x2 ≥ 5; 2x1 - x2 ≥ 12; x1 + 3x2 ≥ 4; x1 ≥ 0; x2 unrestricted.
3) Max. z = 10x1 + 20x2; s.t. x1 + x2 = 4; 2x1 - 3x2 ≤ 7; x1, x2 ≥ 0
4) Max Z = x1 - x2 + 3x3; s.t. x1 + x2 + x3 ≤ 10; 2x1 - x3 ≤ 2; 2x1 - 2x2 + 3x3 ≤ 6; x1, x2, x3 ≥ 0
5) Min Z = 4x1 + 6x2 + 18x3; s.t. x1 + x2 ≥ 3; x2 + 2x3 ≥ 5; x1, x2, x3 ≥ 0
6) Max Z = x1 - 2x2 + 3x3; s.t. -2x1 + x2 + 3x3 = 2; 2x1 + 3x2 + 4x3 = 1; x1, x2, x3 ≥ 0
7) Min Z = x1 + x2; s.t. 2x1 + x2 ≥ 4; x1 + 7x2 ≥ 7; x1, x2 ≥ 0
8) Max Z = 2x1 + 5x2 + 3x3; s.t. 2x1 + 4x2 - x3 ≤ 8; - 2x1 - 2x2 + 3x3 ≥ 7; x1 + 3x2 - 5x3 ≥ -2;
4x1 + x2 + 3x3 ≤ 4; x1, x2, x3 ≥ 09
9) Max Z = 3x1 - 2x2 + 4x3 - 6x4; s.t. x1 + x2 + x3 + x4 = 25; 2x1 - 3x2 - x3 + x4 ≤ 40;
7x1 - 3x2 + 4x3 - x4 ≤ 35; 6x1 + 6x2 - x3 + 3x4 ≥ 28; x1, x3 ≥ 0; x2, x4 ≤ 0;
10) Min Z = 4x1 + 3x2 + 6x3; s.t. x1 + x3 ≥ 2; x2 + x3 ≥ 5; x1, x2, x3 ≥ 0
Duality Theorems

Theorem 1. If an optimal feasible solution exists, the optimal values of the primal
problem and the optimal values of the dual problem are identical.
Theorem 2. Suppose that the slack or surplus variable (also called dummy variable)
in the ith constraint of the primal problem is denoted by si and the jth dummy
variable in the dual problem is denoted by tj . If, in the optimum, x∗j > 0, then t∗j = 0;
if y∗i > 0, then s∗i = 0. This means that if a choice variable in the primal problem is
nonzero in the optimum, its corresponding dummy variable in the dual problem must
be zero in the optimum; and if a choice variable in the dual problem is nonzero in the
optimum, its corresponding dummy variable in the primal problem must be zero in
the optimum. This theorem is called complementary slackness property or
complementary slackness condition or complementary slackness principle.
Canonical Form
This method starts from one infeasible extreme point to another attempting to find a
feasible solution. This feasible solution is also the optimal solution. We cannot apply
this method to all LPP unless the following conditions are satisfied.
(1) The LPP must be in the canonical form
(2) All the entries of (Zj - Cj) must be ≥ 0.
Canonical Form of an LPP
(i) The objective function should be of maximization.
(ii) All the constraints should have “≤“ inequality.
(iii) All variables must be “≥ 0”.
If an LPP satisfies the conditions cited above, then it is called in canonical form.
Eg. Rewrite the given LPP in the canonical form Max Z = 3x1+x2 s.t. 2x1 - x2 ≤ 10;
3x1 + 4x2 ≥ 24; x1 ≥ 2; x2 ≥ 0
Dual Simplex Method
Examples
1) Max Z = 3x1 + 2x2 s.t. 2x1 + x2 ≤ 5; x1 + x2 ≤ 3; x1, x2 ≥ 0
2) Max Z = -2x1 - 9x2 - 24x3 - 8x4 - 5x5; s.t. x1 + x2 + 2x3 - x5 - x6 = 1; -2x1 + x3 + x4 +
x5 - x7 = 2; xj ≥ 0 (for j = 1, 2, ..., 7)
3) Max Z = 2x1 + x2; s.t. x1 + 2x2 ≤ 10; x1 + x2 ≤ 6; x1 - x2 ≤ 2; x1 - 2x2 ≤ 1; x1, x2 ≥ 0
4) Min Z = 3x1 + x2; s.t. x1 + x2 ≥ 1; 2x1 + 3x2 ≥ 2; x1, x2 ≥ 0
5) Min Z = x1 + 2x2 + 3x3 + 4x4; s.t. x1 + 2x2 + 2x3 + 3x4 ≥ 30; 2x1 + x2 + 3x3 + 2x4 ≥ 20;
xj ≥ 0 (for j = 1, 2, 3, 4)
6) Min Z = 6x1 + 7x2 + 3x3 + 5x4; s.t. 5x1 + 6x2 - 3x3 + 4x4 ≥ 12; x2 + 5x3 - 6x4 ≥ 10; 2x1
+ 5x2 + x3 + x4 ≥ 8; xj ≥ 0 (for j = 1, ..., 4)
7) Max Z = -3x1 - 2x2; s.t. x1 + x2 ≥ 1; x1 + x2 ≤ 7; x1 + 2x2 ≥ 10; x2 ≤ 3; x1, x2 ≥ 0
8) Min Z = x1 + 2x2 + 5x3; s.t. | - x1 + 10x2 - 3x3| ≤ 15; 2x1 + x2 + x3 ≥ 10; x1, x2, x3 ≥ 0
9) Min Z = 2x1 + 2x2; s.t. 2x1 + 4x2 ≥ 1; x1 + 2x2 ≥ 1; 2x1 + x2 ≥ 1; x1, x2 ≥ 0
10) Max Z = -3x1 - 2x2; s.t. x1 + x2 ≥ 1; x1 + x2 ≤ 7; x1 + 2x2 ≥ 10; x2 ≤ 3; x1, x2 ≥ 0
11) Max Z = 5x1 - 2x2 + 3x3; s.t. 2x1 + 2x2 - x3 ≥ 2; 3x1 - 4x2 ≤ 3; x2 + 3x3 ≤ 5; x1, x2, x3
≥0
Mathematical Programming
1Transport Problem
Transport Problem
• m origins (production centres/warehouses) and n destinations (market places) in a
transportation system.
• The origin i (i = 1, 2, ..., m) has si (si > 0) units of single product of supply, whereas
the destination j (j = 1, 2, ..., n) requires the demand (dj > 0) dj units.
• Cost of transporting the unit from the origin i to the destination j is cij.
• It is required to determine the number of units of the product to be transported from
origin i to destination j, for all i and j, so that the total transportation cost can be
minimized, subject to the supply and demand constraints.
• Let xij (xij ≥ 0) denotes the number of units transported from the ith origin to jth
destination for all i = 1, 2, ..., m and j = 1, 2, ..., n.
• The relationship between the different elements of a transportation problem can be
given in a tabular form. The box (i, j) will be called cell (i, j).
Linear Programming form

• If the condition is satisfied for a transportation model, then it is called a balanced

transportation problem. Otherwise, the problem is called unbalanced transportation
problem. Even an unbalanced model can be converted to a balanced one.
Solutions
• Basic and Non-basic Cells. If xij be the allocation to the cell (i, j) and xij > 0 is called
positive allocation and the cell (i, j) is called occupied/basic cell. When xij = 0 for
any cell (i, j), it is named as non-basic cell (the value need not be represented in that
cell).
• Feasible Solution. In a transportation problem, a solution is called feasible if the set
of xij values (xij > 0) satisfies all the supply and demand requirements.
• Basic Feasible Solution. If a feasible solution containing exactly (m + n − 1)
basic/occupied cells then it is called basic feasible solution (non-degenerate
solution).
• Degenerate Basic Feasible Solution. If a feasible solution containing less than (m +
n - 1) basic cells, then it is called degenerate basic feasible solution (degenerate
solution).
• Optimal Solution. A basic feasible solution that minimizes (or maximizes the profit)
the total transportation cost is called optimal solution.
Solutions

• TP - Transportation Problem • MODI - Modified Distributive

• S - Supply corresponding to the Method
origin • M - Number of Origins (rows)
• TS - Total Supply • N - Number of Destinations
• D - Demand corresponding to the (columns)
column • A - Actual Allocation of Units
• TD - Total Demand • TC - Total Transportation Cost
• NWCR - North West Corner Rule • RS - Remaining Supply
• RMM - Row Minima Method • RD - Remaining Demand
• CMM - Column Minima Method • VN - Vogel Number
• LCM - Least Cost Method • ε - The Least Positive Quantity (10-6)
• VAM - Vogel’s Approximation • K - Number of Basic Cells
Method
Solutions
Methods to find the initial basic feasible
solution
1. North West Corner Rule (NWCR)
2. Row/Column Minima Method (RMM)
3. Least Cost Method or Matrix Minima Method (LCM/MMM)
4. Vogel’s Approximation Method (VAM)
North-West conner rule
Examples

1) Write down the initial basic feasible solution to the 1 st 1)

transportation problem by NWCR
2) Convert the unbalanced transportation problem (the
2nd one) into a balanced one.
3) Find an initial basic feasible solution of the 3 rd
problem using NWCR 2)
3)
Row/Column minima method
Examples

1) Solve the 1st transportation problem using row minima

method.
2) Solve the 2nd transportation problem using column
minima method
Least Cost/Matrix Minima Method
Examples

Obtain an initial basic feasible solution to the 1)

following transportation problem using
matrix minima method.

2)
Vogel’s Approximation Method (VAM)

• N - Number
• CL - Close
• VN - Vogel Number (penalty cost)
• Rm - mth Row
• RVN - Row Vogel Number
• Cn - nth Column
• CVN - Column Vogel Number
• UG - Units Given
• R1 - 1st Row
• C1 - 1st Column
• SC - Selected Cell
Examples

1) Find the initial basic feasible solution of 1)

the following transportation problem (1st
table) by VAM.
2) Solve the transport problem (2nd table)
3) The following table (3rd table) shows all
the information regarding the available
supply at each warehouse, the requirements
of each market and the unit transportation
cost from each warehouse to each market.
Find the initial solution using VAM.
Stepping Stone Method/Modified
Distributive Method
• NBC - Number of Basic Cells
• I - Number of Iterations
• BCs - Basic Cells
• VU - Vertically Up

• VD - Vertically Down
• HR 0 Horizontally Right
• HL - Horizontally Left
• R - Row
• C - Column
• H - Horizontal
• V - Vertical
Examples

1) Solve the following transport problem:

(a) (b)
2) Consider the transportation problem. Find (i) initial solution using
NWCR. (ii) Find the optimum solution. Is the solution unique? If not find
the alternate solution.
Examples

1)

2)
Mathematical Programming
1Assignment Problem
Assignment Problem

Consider n jobs and n persons. Assume that each job can be done only by one person
and the expense a person required for completing the ith job (i = 1, 2, ..., n) by the jth
person (j = 1, 2, ..., n) is denoted by a real number Cij. On the whole, this model deals
with the assignment of n candidates to n jobs in such a way that to minimize the time
(cost) for over all completion of the jobs.

Obviously, the assignment problem is a special case of the standard transportation

problem with the following properties:
(i) The total number of origins is equal to the number of total destinations.
(ii) Each origin is represented by one job, that is, si = 1, where i = 1, 2, ..., n.
(iii) Each destination is represented by one person, that is, dj = 1, where j = 1, 2, ..., n.
Hungarian Algorithm
Examples

1) 2) 3)

4) A manager has four salesmen and four sales districts. He estimates that the profit per
day from each salesman in each district would be as follows. Find the assignment of
salesmen to districts that will result in maximum profit.
Examples

5) Solve the following problem to maximize the profit. Which job is to be left?

6) Impossible Assignment