PART III LINEAR PROGRAMMING- Solving LPP= By

Simplex Method
1

 For LPPs with several decision variables (if they are more than two),
two we may
not be able to graph the feasible region.

 If so, an algorism which is called a Simplex Method, will be an appropriate to

solve the LLP instead of using the graphical method.

 The simplex method examines the extreme points in a systematic manner,

 repeating the same set of steps of the algorism until an optimal solution is
reached.
 It is for this reason that it is also called the iterative method.
method
 2

 Linear Programs in Standard Form

 The use of the simplex method to solve a LPP requires that the problem be converted into its

“Standard" form.

Characteristics of the standard form of the LPP-

a) All the constraints should be expressed as equations by adding slack or surplus and/or artificial

variables.

b) The right-hand side of each constraint should be made non-negative; if not, this should be done by

multiplying both sides of the resulting constraint by -1.

 3

 Linear Programs in Standard Form

 For your reference, the standard for of the LPP is expressed as:

Max c1x1 + c2x2 + …+ cnxn + 0s1 + 0s2 + … +0sm

subject to :
a11x1 + a12x2 + …+ a1nxn +s1= b1

am1x1 + am2x2 +…+ amnxn + s2= bm

x1 ; … xn, s1, s2 ≥ 0
where the objective is maximized, the constraints are equalities and the variables
are all nonnegative.
 4

 This is done as follows:

 If the problem is min z, convert it to max -z.

 If a constraint is ai1x1+ai2x2++ainxn ≤ bi, convert it into an equality constraint by
adding a nonnegative slack variable si.
 The resulting constraint is ai1x1+ai2x2+…+ainxn+si = bi,

where si ≥ 0.

 If a constraint is ai1x1 + ai2x2 + … + ainxn ≥ bi, convert it into an equality

constraint by subtracting a nonnegative surplus variable si.
 The resulting constraint is ai1x1+ai2x2+…+ ainxn - si = bi,

where si ≥ 0.
 5
Steps for Simplex Tabular Form

1. Convert the LP to standard form

 i.e. convert constraint inequality into equality form by introducing a variable

called slack variable.
 A slack variable (s) is added to the left hand side of a ‘≤’ to convert the
constraint inequality into equality.
 The value of the slack variable represents unused resource or idle capacity.
 A slack variable emerges when the LPP is a maximization problem.
 Thus, they don’t produce any product and their contribution to profit is zero
 Slack variables are added to the objective function with zero coefficients.
 6

Steps for Simplex Tabular Form

2. Obtain a basic feasible solution (bfs) or initial simplex tableau from the
standard form

3. Determine whether the current bfs is optimal. If it is optimal, stop.

4. If the current bfs is not optimal, determine which non-basic variable should

become a basic variable and which basic variable should become a non-basic
variable to find a new bfs with a better objective function value.

5. Go back to step 3.
7
 Step 4 construct the initial tabula

8
Step 4 construct the initial tabula
 9

 Step 5 choose the entering variable with maximum positive number in cj-zj
row
 Step 6 choosing the leaving variable ration RHS
10
11
 Minimization Problem

12
 MINIMIZATION PROBLEMS
 Minimize Z with inequalities of constraints in “> “form
 There are two methods to solve minimization LP problems:
 1. Direct method/Big M-method/
 Using artificial variables

 2. Conversion method
 Minimization by maximizing the dual

 Surplus Variable (-s):

 A variable inserted in a greater than or equal to constraint to create equality.
It represents the amount of resource usage above the minimum required
usage.
 Surplus variable is subtracted from a > constraint in the process of
converting the constraint to standard form.
 Neither the slack nor the surplus is negative value. They must be positive or
zero.
 13
 Example:
 2x1+x2 < 40 ==>is a constraint inequality
 x1= 12 and x2= 11==> 2x1+x2+s = 40 ==>2(12)+11+s = 40
 ==> s=5 unused resource
 5x1+3x2 < 93
 x1= 12 and x2= 11==> 5x1+3x2+s = 93 ==>5(12)+3(11)+s = 93 
 ==> s=0 unused resource (No idle resource)
 2x1+x2 >40
 x1= 0 and x2= 0 ==> 2x1+x2- s = 40 ==>2(0)+(0)-s = 40
 ==> s= -40
(This is mathematically unaccepted)
 Thus, in order to avoid the mathematical contradiction, we have to add
artificial variable (A).
 14

 Artificial variable (A):

 Artificial variable is a variable that has no meaning in a physical sense
but acts as a tool to create an initial feasible LP solution. 
 Note:
 Type of constraint To put into standard
form
< --------------------------------------------- Add a slack variable
= ---------------------------------------------Add an artificial variable
> ---------------------- Subtract a surplus variable and add artificial
variable
 The Big M-method

15
The Big M-method
 /Charnes Penalty Method/
 The Big-M Method is a method which is used in removing artificial
variables from the basis .In this method; we assign coefficients to artificial
variables, undesirable from the objective function point of view.
 If objective function Z is to be minimized, then a very large positive price
(called penalty) is assigned to each artificial variable.
 Similarly, if Z is to be maximized, then a very large negative price (also
called penalty) is assigned to each of these variables.
 Following are the characteristics of Big-M Method:
 High penalty cost (or profit) is assumed as M
 M is assigned to artificial variable A in the objective function Z.
 Big-M method can be applied to minimization as well as maximization
problems with the following distinctions:
 1. Minimization problems

16
 -Assign +M as coefficient of artificial variable A in the objective function Z
Maximization problems:

 -Here –M is assigned as coefficient of artificial variable A in the objective
function Z
 Coefficient of S (slack/surplus) takes zero values in the objective function Z
 For minimization problem, the incoming variable corresponds to the
highest negative value of Cj-Zj.
 Solution is optimal when there is no negative value of Cj-Zj.(For
minimization case)
 Example:
 1. Minimize Z=25x1 +30x2
 Subject to:
 20x1+15x2 > 100
 2x1+ 3x2 > 15
 x 1, x2 >0
 17
 Solution
 Step 1
 Standardize the problem
 Minimize Z=25x1 +30x2 +0s1+0s2 +MA1+MA2
 Subject to:
 20x1+15x2- s1+A1 = 100
 2x1+ 3x2 –s2+A2 = 15
 x1, x2 , s1, s2 ,A1 ,A2 > 0  
 Step 2
 Initial simplex tableau
 The initial basic feasible solution is obtained by setting x1= x2= s1= s2=0
 No production, x1= x2= s1=0==>20(0) +15(0) - 0+A1 = 100 ==> A1 = 100
 x1= x2= s2=0==>0(0)+3(0) - 0+A2 =15==> A2 = 15
18
19
20
 Simplex Problem (Mixed
Constraints)
 Simplex Problem (Mixed Constraints) 21

 The following points may be noted in this regard

 A situation may arise when the constraints are ≥ type or ≤ type or =

 Introduce slack variable incase of ≤ type constraint.

 Introduce surplus variables and artificial variables in case of ≥ type constraint

 Introduce artificial variables if the constraint is = type

 Assign zero co-efficient to the slack variables and surplus variables in the objective function

 Assign +M in case of minimization and –M incase of maximization problems to the artificial

variables in the objective function.

 In case of maximization, stop where Cj – Zj row contains zero or negative elements

 In case of minimization, stop where Cj – Zj row contains zero or positive elements


 22

Exercise
Minimize Z=60x1 + 80x2
St
X2 ≥ 200
X1≤400
X1 + X2 =500
X1, X2 ≥ 0

 Assignment for group 1 (15%)
23
 An Air force is experimenting with three type of bombs P, Q and R in which

three kinds of explosive Viz. A, B and C will be used. Taking the various factors
in to consideration, it has been decided to use the maximum 600 KG of
explosive A, at least 480 KG of explosive B and exactly 540 of explosive C.
bomb P requires 3, 2, 2 KG of A, B and C respectively. Bomb Q requires 1, 4, 3
KG of A, B and C respectively. Bomb R requires 4, 2, 3 KG of A, B and C
respectively. Now bomb P will give the equivalent of 2 tons of explosive, bomb
Q will give 3 tone explosive and bomb R will give 4 tone explosive. Under what
production schedule can the Air force make the biggest bang?


 Assignment for group 2 (15%)
24

Solve the following LPP: ( using simplex algorism)

Max. Z = 4x1 + 5x2 -3x3
Subject to x1 + x2 + x3 = 10
 x1 – x 2 ≥ 1
 2x1 + 3x2 + x3 ≤ 30
Where x1, x2, x3 ≥ 0
 Assignment for group 3 (15%)
25
 A firm produces three products A, B and C, each of which passes
through three departments: Fabrication, finishing and packaging. Each
unit of product A requires 3, 4 and 2 hours; a unit of B requires 5, 4
and 4 hours while each unit of product C requires 2, 4, 5 hours
respectively in the three departments. Every day, 60 hours are
available in the fabrication department, 72 hours in the finishing
department and 100 hours in the packaging department. If the unit
contribution of product A is $ 5, of product B is $ 10, and of product C
is $ 8. Required
a. The number of units of each of the products that should be made each
day to maximize the total production. ( using simplex algorism)
b. Determine if any capacity would remain unutilized.
 Assignment for group 4 (15%)
26

A company produces two types of pen, say A and B. Pen A is a

superior quality and pen B is lower quality. Profit on pens A
and B is $ 5 and $ 3 per pen respectively. Raw material
required for each pen A is twice as that for pen B. The supply
of raw materials is sufficient only for 1000 pens of type B per
day. Pen A requires a special clip and only 400 such clips are
available per day. For pen B, only 700 clips are available per
day. Required
a. Formulate a LP model for the problem.
b. Find graphically the product mix so that the company can
maximize profit.
 Duality
27

 Associated with any LPP is another LPP called its dual.

 The first way of stating linear programming problem is called the primal of the problem.

 The second way of stating the same problem is called dual.

 Each LP maximizing problem has corresponding dual minimizing problem.

 Similarly each LP minimizing problem has its corresponding dual maximizing problem.

 Optimal solution for the primal and the dual are equivalent but they are derived through

alternative procedures.
 Formulating the dual problem from a given primal is not difficult, and once it is

formulated, the solution procedure is exactly the same as for any LP.
 Formulation of the Dual Problem
28

Primal problem

Max c1x1 + c2x2 + …+ cnxn

 subject to :
 a11x1 + a12x2 + …+ a1nxn ≤ b1
 a21x1 + a22x2 + … + a2nxn≤ b2
 : :
am1x1 + am2x2 +…+ amnxn ≤ bm
 x1, x2; … xn, ≥ 0
 Cont’d
29

The Dual of the problem

Min. Z = b1y1 + b2y2 +….+ bmyn
 ST.
a11y1 + a21y2 + … + am1ym ≥ c1
a12y1 + a22y2 + … + am2ym ≥ c2
 : :
a1ny1 + a2ny2 + … + amnym ≥ cn
 
y1, y2, ….., ym ≥ 0
 Rules for constructing the Dual problems
30
1. if the objective of the one problem is to maximize, the
objective of the other is to minimize.
2. the maximization problem should have all ≤ constraints and
the minimization problems should have ≥constraints. If in a
maximization case any constraints is ≥ type it can be multiplied
by -1 and sign got reversed. The same is true in minimization.
3. the number of primal decisions variables (xi) equals the
number of dual constraints and the number of primal
constraints equals the number of dual variables (yi). If the
primal problem has “m” constraints and “n” variables then dual
problem will have “n” constraints and “m” variables.
 Cont’d
31
4. the cj of the primal in the objective function appears as “b”
constant of the dual constraint, and the primal “b” constant
appears as unit contribution rate i.e. Cj of the dual decision
variable in the objective function.
Min. Z = 40x1 + 200x2
ST
 4x1 + 40x2 ≥ 160
 3x1 + 10x2 ≥ 60
 8x1 + 10x2 ≥ 80
 X1, x2 ≥ 0