Vishal Mishra (IBS, Hyderabad)

Linear Programming
Linear programming problem (LPP): Maximization Problem

Example: A start-up pet food bakery has two key bakery items to offer to their customers i.e.,
vegetarian cookies and non-vegetarian cookies. The production of these two items requires
processing in two key sections: Kitchen and Packaging. One kg of vegetarian cookies requires 2
hrs in preparation (i.e., in kitchen) and 60 minutes in packaging. While, one kg of non-
vegetarian cookies requires 2.5 hrs in kitchen and 30 minutes in packaging. The total time
available is 35 hours per week in kitchen and 15 hours per week in packaging section. The profit
per kg of vegetarian cookies is INR 200 and for a kg of non-vegetarian cookies is INR 225*.
What is the quantity of each type of cookies that this start-up should produce in a week in
order to maximize its profits?

*Assume that all the cookies get sold.

 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem

Objective Function:
Maximize 200 V + 225 N

Such that (constraints),

2 V + 2.5 N <= 35
1 V + 0.5 N <= 15
V, N >= 0
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem

Objective Function:
Maximize 200 V + 225 N

Such that (constraints),

2 V + 2.5 N <= 35
1 V + 0.5 N <= 15
V, N >= 0
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem
Optimal Solution
Objective Cell (Max)
Cell Name Original Value Final Value
$J$9 Unknowns OF 0 3416.666667

Variable Cells
Cell Name Original Value Final Value Integer
$G$9 Unknowns V 0 13.33333333 Contin
$H$9 Unknowns N 0 3.333333333 Contin

Constraints
Cell Name Cell Value Formula Status Slack
$I$12 C1 35 $I$12<=35 Binding 0
$I$13 C2 15 $I$13<=15 Binding 0
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem
Sensitivity Analysis: How sensitive is the optimal solution point and optimal solution value to
the changes in

1. Objective Function Coefficients &

2. RHS of the constraints (capacity/resource availability)

Ideally one needs to solve the problem again

But sometimes 1). One need not solve the problem again and/or 2) Find out the impact of the
above changes from some of the output values (that excel-solver provides).
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem: Sensitivity Analysis: Objective Function Coefficients
What happens to the optimal solution (point) when the objective function coefficient changes:

Note that objective function itself can be seen as an equation of a straight line. One can write
Maximize K = 200 V + 225 N
OR
N = K/225 – 200V/225

(Similar to Y = C + m X
OR yො = b0 + b1 X )
Note: Slope

Graphical Illustration
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem
Sensitivity Analysis: Objective Function Coefficients
What happens to the optimal solution (point) when the objective function coefficient changes
Individual Change:
Variable V: Current value of the coefficient (profit contribution) is 200.
The optimal solution point (13.33, 3.33) will not change for the range of the
coefficient of V between 180 to 450 (based on allowable increase/decrease),
keeping all the other aspect of the problem constant.
Variable Cells
Final Reduced Objective Allowable Allowable
Cell Name Value Cost Coefficient Increase Decrease
$G$9 Unknowns V 13.33333333 0 200 250 20
$H$9 Unknowns N 3.333333333 0 225 25 125
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem
Sensitivity Analysis: Objective Function Coefficients
What happens to the optimal solution (point) when the objective function coefficient changes
Individual Change:
Variable N: Current value of the coefficient (profit contribution) is 225.
The optimal solution point (13.33, 3.33) will not change for the range of the
coefficient of N between 100 to 250 (based on allowable increase/decrease),
keeping all the other aspect of the problem constant.
Variable Cells
Final Reduced Objective Allowable Allowable
Cell Name Value Cost Coefficient Increase Decrease
$G$9 Unknowns V 13.33333333 0 200 250 20
$H$9 Unknowns N 3.333333333 0 225 25 125
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem
Sensitivity Analysis: Objective Function Coefficients
Simultaneous Change: (both the coefficients change): 100% Rule
The optimal solution point (13.33, 3.33) will not change so long as the total
percentage change (coefficients of both the variables together) is <= 100%,
keeping all the other aspect of the problem constant. Note that the optimal
solution may change if the total % change is more than 100%.

Example: If coefficient of V increase by 50 and that of N decreases by 25

% increase (V): (50/250)*100 = 20
% decrease (N): (25/125)*100 = 20

Total % (increase/decrease): 40
So, in this case the optimal solution point will remain same (i.e. 13.33, 3.33).
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem: Sensitivity Analysis: Constraints (RHS)
What happens to the optimal solution (and the value i.e. objective function value) when the
RHS of the constraint (resource availability) changes: One needs to solve the problem again as
there may be a different optimal solution (R’ in the figure below)

Constraint 1 written as
Equality i.e.,
2 V + 2.5 N = 35 OR
N = 35/2.5 – 2V/2.5

Note: Intercept

Graphical Illustration
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem: Sensitivity Analysis: Constraints (RHS)
What happens to the optimal solution (and the value i.e. objective function value) when the
RHS of the constraint (resource availability) changes
Individual Change:
Constraint 1: Current value of the RHS of the constraint is 35. The shadow price is the amount
by which the optimal solution value will change (INR 83.33) for a unit change in the
RHS of the constraint 1 (i.e. 1 hr), within the allowable increase/decrease range i.e.
between 30 and 75 hr), keeping all the other aspect of the problem constant (the
optimal solution point may also change). Note that the shadow price may change
beyond the aforementioned range of the constraint 1. So if 1 more hr of kitchen time
is made available (36 hrs) then profit will increase by INR 83.33 (added to 3416.67).
Constraints
Final Shadow Constraint Allowable Allowable
Cell Name Value Price R.H. Side Increase Decrease
$I$12 C1 35 83.33333333 35 40 5
$I$13 C2 15 33.33333333 15 2.5 8
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem: Sensitivity Analysis: Constraints (RHS)
What happens to the optimal solution (and the value i.e. objective function value) when the
RHS of the constraint (resource availability) changes
Individual Change:
Constraint 2: Current value of the constraint is 15. The shadow price is the amount by which
the optimal solution value will change (INR 33.33) for a unit change in the RHS of the
constraint 1 (i.e. 1 hr), within the allowable increase/decrease range i.e. between 7
and 17.5 hr), keeping all the other aspect of the problem constant (the optimal
solution point may also change). The shadow price may change beyond the
aforementioned range of the constraint 2. So, if 1 more hr of packaging time is made
available (16 hrs) then profit will increase by INR 33.33 (added to 3416.67).
Constraints
Final Shadow Constraint Allowable Allowable
Cell Name Value Price R.H. Side Increase Decrease
$I$12 C1 35 83.33333333 35 40 5
$I$13 C2 15 33.33333333 15 2.5 8
 Vishal Mishra (IBS, Hyderabad)

Linear Programming: Excel-Solver

Maximization Problem: Sensitivity Analysis: Constraints (RHS):
Simultaneous Change (100% rule)
The shadow prices will not change so long as the total percentage change (RHS of all the constraints) is
<= 100%, keeping all the other aspect of the problem constant. Note that the shadow price and the
optimal solution may change if the total % change is more than 100%. Also note that the problem will
have to be solved again to identify the optimal solution point (values of V & N).

Example: If RHS of C1 (hrs in kitchen) increases by 10 and that of C2 (hrs in packaging) decreases by 4.
% increase (C1): (10/40)*100 = 25
% decrease (C2): (4/8)*100 = 50
Total % (increase/decrease): 75
Since this is less than 100%, the shadow price do not change. Thus the new optimal solution value will
be 3416.66 + (10*83.33) – (4*33.33) = 4116.67 (of course, at a new optimal solution point other than R
(13.33, 3.33)).