Scribd
Upload
10 views28 pages

Operations Research Session 7-8

Uploaded by

sahil.mp2026
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views28 pages

Operations Research Session 7-8

Uploaded by

sahil.mp2026
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 28

OPERATIONS RESEARCH

IIM Bodh Gaya, MBA 2024-2026, Term II – 07-08

Dr. Rohit Agrawal


Email:- [email protected]

Problem: Solve using Simplex


Minimize Cost: Z = 4��1 +

��2 3��1 + ��2 ≥ 3

4��1 + 3��2 ≥ 6

��1 + 2��2 ≤ 4

X, Y ≥ 0

2
Solution

3
BIG M Method

The problem is converted to canonical form by adding slack, surplus and artificial variables as

appropriate. 1. As the constraint-1 is of type '=' we should add artificial variable A1

2. As the constraint-2 is of type '≥' we should subtract surplus variable S1 and add artificial variable

A2 3. As the constraint-3 is of type '≥' we should subtract surplus variable S2 and add artificial

variable A3
4

Solution: Minimization
Cj

RHS
5

Equations for iteration 2


6

Equations for iteration 3


7

Equations for iteration 4


8

Results from iteration 4


9

DUALITY
Try and get an UPPER ESTIMATE of Z without solving

Maximize Z 10X1 + 9X2


Subject to
3X1 + 3X2 ≤ 21
4X1 + 3X2 ≤ 24
X1, X2 ≥ 0
We are interested in finding out, what would be the upper estimate of the objective
function without actually solving it.

Now, let us take the first constraint 3X1 + 3X2 ≤ 21, now let us multiply the first
constraint by 10. we will get 30X1 + 30X2 ≤ 210

• This tells us ------------

• We also should have X1, X2 ≥ 0. So, when X1, X2 ≥ 0,

• 30X1 + 30X2 ≥ 10X1 + 9X2, because X1 and X2 ≥0

• Therefore, 10X1 + 9X2 which itself ≤ 30X1 + 30X2 will have to be ≤210

Objective function at the optimum cannot exceed 210


• Same multiply first constraint by 4 and check………..

• Now, between the upper estimates of 210 and 84, 84 is more meaningful, because we can now say that the
value of
the objective function cannot exceed 84 and that’s a more meaningful expression rather than saying, it cannot

exceed 210.
2. Now, let us multiply the second constraint by 3, so we multiply the second constraint by 3 to get 12X1 + 9X2
≤ 72.
Now, again going back, so long as X1 and X2 are greater than or equal to 0, 10X1 + 9X2 ≤12X1 + 9X2 and
since, 12X1 + 9X2 itself should be ≤72.

10 X1 + 9X2 has to be ≤72 and 72 is now an upper estimate to the value of the objective function.

Now, instead of taking one constraint at a time, can we do the same thing with multiple constraints? Now, in this
example there are only two constraints therefore, we take both of them, if the problem had a larger number we
could take a subset.
So, if we take multiple constraints we take two of them. Multiply the first constraint by 3 and multiply the
second constraint by 1 or simply add the second constraint. So, when we multiply the first constraint by 3,
we get 3 x 3. So, what we do is, this is multiplied by 3 and this is multiplied by 1, which is the same as adding it.

13X1 + 12X2 ≤ 87

So, when multiply the first constraint by a and multiply the second constraint by b, the new
constraint for X1 will have 3a + 4b.
Now, a and b have to be such that 3a + 4b ≥ 10 and they also have to be such that, 3a + 3b ≥ 9

If a and b that satisfy these two conditions after multiplication,

then 21a + 24b is an upper estimate of the objective function value.

A. Now, the original problem that we had is called the primal, and the new problem is called
the dual.
B. Now, for every linear programming problem which is called as primal, there is another
linear programming problem which is called the dual.

PRIMAL

DUAL
Primal-Dual
In any LPP, the Dual of the Dual is
relationships the Primal itself
For Primal and Dual
problems

For Constraints
and Variables

PRIMAL
PRIMAL DUAL
PRIMAL DUAL
Every feasible
solutions
Duality theorems
Weak Duality Theorem
• For every maximization primal, every feasible solution to the dual has
an objective function value greater than or equal to every feasible
solution to the primal.

Optimality Criterion Theorem


• If the primal and dual have feasible solution with the same value of
objective function, then both have optimal solutions with same value of
objective function.

Practice for duality


24
Thank you
Any discussion is Welcome!

You might also like