Master of Business Administration: Quantitative Analysis For Management Decisions
Master of Business Administration: Quantitative Analysis For Management Decisions
Topics To Be Covered
Formulation of LPP
Graphical method for two variables
Solution- BFS. Optimal , degenerate, non degenerate
solutions
Simplex method
Big ‘m’ method
2
1.1. Formulation of LPP
Linear Programming is a mathematical process that as bee
developed to help management in decision making involving
the efficient allocation of scares resources to achieve a
certain objectives.
LP is a class of mathematical programming model which
deals with the program of allocation limited resource among
competing activities in an optimum manner.
5
Mathematical Form of Linear Programming
Let the decision (input) variables be X1, X2, …….Xn and m
as the set of constraint. Both decision variables and
constraints are linear function of the variable.
The standard form of a general linear programming may be
written in the following form.
6
Con’t…
Subject to constraints
𝑎11 𝑥1 + 𝑎12 𝑥2 + ⋯ … … … … . . +𝑎1𝑛 𝑥𝑛 ≤, =, ≥ 𝑏1
𝑎21 𝑥1 + 𝑎22 𝑥2 + ⋯ … … … … . . +𝑎2𝑛 𝑥𝑛 ≤, =, ≥ 𝑏2
.
.
𝑎𝑚1 𝑥1 + 𝑎𝑚2 𝑥2 + ⋯ … … … … . . +𝑎𝑚𝑛 𝑥𝑛 ≤, =, ≥ 𝑏𝑚
where x1, x2,…..xn are decision (input) variables, c1, c2,…cn are called
constant are representing cost or profit coefficients i.e. per unit contribution
to the objective function, aij (i=1,2,….m and j=1,2,…..n) are called
structural coefficients, b1, b2,……,bm represents an availability of m
constraints and xj 0 simply implies that the Zi’s must be non-negative.
7
Assumptions in linear programming models
11
The Graphical Methods
12
Con’t…
corner points.
optimal value.
14
Example 1
Solution:
16
Con’t…
Point X1 X2
O 0 0 0
A 0 10 300
D 12 0 240
18
Example 2:
19
Con’t…
Solution
1. Formulation of mathematical modeling of LPP
Max Z=300X1 +250X2
St:
2X1 +X2< 40
X1 +3X2< 45
LPP Model
X1 < 12
X1, X2 >0
20
Con’t…
2. Convert constraints inequalities into equalities
2X1 +X2 = 40
X1 +3X2= 45
X1 = 12
3. Draw the graph by intercepts
2X1 +X2 = 40 ==> (0, 0) and (20, 0)
X1 +3X2= 45==> (0, 15) and (45, 0)
X1 = 12==> (12, 0)
X1, X2 >0
21
Con’t…
22
Con’t…
7. Interpret the result
Corner Points Coordinates Max Z=300 X1 +250X2
A (0,0) $0
B (0,15) $3750
C (12,11) $6350
D (12,0) $3600
Interpretation:
12 units of product A and 11 units of product B should be
produced so that the total profit will be $6350.
23
Example 3:
_____________________________________________________________________
Min.Z 25 X 130 X 2
St :
20 X 115 X 2 100 LPP Model
2 X 13 X 2 15
X1, X 2 0
25
Con’t…
Constraint
equation:
20X1 +15X2=100 ==> (0, 20/3) and (5, 0)
2X1+3X2=15 ==> (0, 5) and (7.5, 0)
X1, X2 >0
26
Con’t…
X1 =2.5
X2=3.33 and
Min Z= 162.5
27
SPECIAL CASES IN GRAPHICS METHODS
1. Redundant Constraint
If a constraint when plotted on a graph doesn’t form part of
the boundary making the feasible region of the problem that
constraint is said to be redundant.
If the removal of constraints of a LPP does not alter the
optimal solution to the problem, the constraint is said to be
redundant constraint. In other words, constraint appears
redundant when it may be more binding/restrictive/ than
other.
28
Con’t…
Consider the following maximization function
Max.Z 40 X 135 X 2
St :
2 X 13 X 2 60
4 X 13 X 2 96
4 X 1 3.5 X 2 105
X1, X 2 0
The feasible and optimal solutions were found to be
31
Con’t…
Example:
The information given below is for the products A and B.
_____________________________________________________________________
Machine hours per week Maximum available
Department Product A Product B per week
_____________________________________________________________________
Cutting 3 6 900
Assembly 1 1 200
Profit per unit $8 $16
_____________________________________________________________________
33
Con’t…
Interpretation:
Both C and D are optimal solutions. Any point on the line
segment CD will also lead to the same optimal solution.
==>Multiple optimal solutions provide more choices for
management to reach their objectives.
34
3. Infeasible Solution
A solution is called feasible if it satisfies all the constraints and
the constraints and non-negativity condition. However, it is
sometimes possible that the constraints may be inconsistent so
that there is no feasible solution to the problem. Such a
situation is called infeasibility.
Example: Consider the following LPP
Note:
37
Con’t…
38
SIMPLEX METHOD
39
Con’t…
42
Con’t…
.
.
.
44
Con’t…
In matrix form the standard form is expressed as:
and where is the row vector, are column vectors
45
Con’t…
To employ the simplex algorithm, we use the following
tableau as a device.
𝑐𝑗 𝑐1 𝑐2 … 𝑐𝑛 0 0… 0
Coefficients of Basic Basic 𝑥1 𝑥2 … 𝑥𝑛 𝑠1 𝑠2 … 𝑠𝑚 b 𝑏ൗ
𝑎𝑖𝑗
Variables in the Variable
solution of 𝑐𝑗
0 𝑠1 𝑎11 𝑎12 … 𝑎1𝑛 1 0… 0
0 𝑠2 𝑎21 𝑎22 … 𝑎2𝑛 0 1… 0
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
0 𝑠𝑚 𝑎𝑚 1 𝑎𝑚 2 … 𝑎𝑚𝑛 0 … 1
𝑍𝑗 0 0 … 0 0 … 0
𝑐𝑗 − 𝑍𝑗
46
Summary of steps in solving a maximizing LP
50
Con’t…
The variable heading that column is which should enter the solution. This
if more than one variable appears with the same maximum value, any one
51
Con’t…
The current basic variable, then, to be replaced and is called leaving
(outgoing or departing) variable and is to be made zero. The row so marked
is called the key (pivot) row. Zero is considered as non-negative and
negative ratios are discarded. The element lying at the intersection of key
column and key row is called key (or pivot) element.
III. Evaluating (updating) the new solution: the new simplex table will be
updated by using Gauss-Jordan computations procedures to find new raw
values by performing some mathematical operations
Pivot Row
a) Replace the leaving variable in the basic column with the entering
variable.
b) New pivot row = Current pivot Pivot element
All other rows, including z
a) New row = (Current row value) (its pivot column coefficient) New
pivot row
52
Con’t…
Step
7: Perform optimality test for the new Simplex table
Compute = and also the row, the algorithm terminates if
all the row values are either zero or negative, otherwise the
algorithm continues in the same fashion.
53
Example 1:
54
Soution:
Available capital=5000 birr
Available land =100 acres
Crops cost/acres profit
Potatoes 100 30
Corns 40 10
Cabbage 70 15
55
Soution:
Subject to:
Subject to:
56
Con’t…
The initial simplex tableau is:
Cj 30 10 15 0 0
Var. in Var. in Cj 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 b b/aij
soln.
𝑠1 0 100 40 70 1 0 5000 50
𝑠2 0 1 1 1 0 1 100 100
𝑍𝑗 0 0 0 0 0
𝑐𝑗 − 𝑍𝑗 30 10 15 0 0
57
Con’t…
The second simplex tableau is:
R1 (new) R1(old)/100 1, 0.4, 0.7, 0.01, 0, 50
R2 (new) R2(old)-1*( R1 (new))0, 0.6, 0.3, -0.01,1, 50
Cj 30 10 15 0 0
Var. in Var. in 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 b b/aij
soln. Cj
𝑥1 30 1 0.4 0.7 0.01 0 50
𝑠2 0 0 0.6 0.3 -0.01 1 50
𝑍𝑗 30 12 21 0.3 0 1500
𝑐𝑗 − 𝑍𝑗 0 -2 -6 -0.3 0
58
Con’t…
All values are less than or equal to zero, thus we cannot
improve the objective function. Value of the variable X1=50
and S2=50 in maximize the objective function. X1 represents
the unused capital portion of land. Therefore the solutions
are X1=50, S2=50, X2, X3, S1=0. The farmer should plant 50
acres of land of potato in order to maximize profit and he
has to leave the other 50 acres of land unplanted and the
maximum profit is 1500 birr.
59
Example 2:
60
Solution:
The information in the problem may be organized using table:
Engineering 1 1 1 100
Labor 10 4 5 600
Administration 2 2 6 300
Profit/100 units 10 6 4
Subject to
Subject to
62
Con’t…
The initial simplex tableau is:
Cj 10 6 4 0 0 0
Var. in Var. in 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 b b/aij
soln. Cj
𝑠1 0 1 1 1 1 0 0 100 100
𝑠2 0 10 4 5 0 1 0 600 60
𝑠3 0 2 2 6 0 0 1 300 150
𝑍𝑗 0 0 0 0 0 0
Cj-Zj 10 6 4 0 0 0
Cj 10 6 4 0 0 0
Var. in Var. in 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 b b/aij
soln. Cj
𝑠1 0 0 0.6 0.5 1 -0.1 0 40 400/6
𝑥1 10 1 0.4 0.5 0 0.1 0 60 600/4
𝑠3 0 0 1.2 5 0 -0.2 1 180 150
𝑍𝑗 10 4 5 0 1 0 600
Cj-Zj 0 2 -1 0 -1 0
64
Con’t…
The
third simplex tableau is:
R1 (new) R1(old)/0.6 0, 1, 5/6, 10/6, -1/6, 0, 400/6
R2 (new) R1(old)-0.4* R1 (new) 1, 0, 1/6, -2/3, 1/6, 0, 100/3
R3 (new) R3(old)-1.2* R1 (new) 0, 0, 4, -2, 0, 1, 100
Cj 10 6 4 0 0 0
Var. in Var. in 𝑥1 𝑥2 𝑥3 𝑠1 𝑠2 𝑠3 b b/aij
soln. Cj
𝑥2 6 0 1 5/6 10/6 -1/6 0 400/6
𝑥1 10 1 0 1/6 -2/3 1/6 0 100/3
𝑠3 0 0 0 4 -2 0 1 100
𝑍𝑗 10 6 40/6 10/3 4/6 0 2200/3
Cj-Zj 0 0 -16/6 -10/3 -4/6 0
Since all Cj-Zj values are not positive, the objective function
cannot be improved. The optimal solution is to obtain the
maximum profit 2200/3 birr.
65
Solving Minimization Problem
66
Con’t…
The minimum value of the objective function is the value
read by the final simplex tableau of the equivalent
maximization problem multiplied by -1.
The second approach requires a change of the rule of
determining the entering variable for a minimization
problem, the variable corresponding to the most negative
will be the next entering variable improved and the
objective function cannot be improved further if all the .
Direct method/Big M-method/Using artificial variables
67
Con’t…
68
Con’t…
◦ Maximization problems:
- Assign–M as coefficient of artificial variable A in the
objective function Z
70
Con’t…
71
Example:
72
Con’t…
M A1 20 15 -1 0 1 0 100 100/20=5
M A2 2 3 0 -1 0 1 15 15/2=7.5
Zj 22M 18M -M -M M M 115 M
Cj - Zj 25 -22M 30- 18M M M 0 0
73
Con’t…
SV X1 X2 S1 S2 A2 Q
74
Con’t…
Cj 25 30 0 0
SV X1 X2 S1 S2 Q
75