Linear Programming
1.  OM applications
2.  Problem formulation
3.  Graphical solution method
4.  Corner point solution method
5.  Using excel
6.  Sensitivity analysis

Linear Programming: related mathematical techniques used to allocate limited resources among competing demands in an optimal way

Applications
1.  Resource allocation
2.  Facility location

Problem formulation
1.  problem characteristics
2.  problem formulation guidelines
3.  two types of problems
4.  exercise

Problem characteristics

Guidelines on model formulation

 
     Requirements >=
     Limitation <=
     Equal =

Two types of problems

Blending (Minimization problem)

Product mix (Maximization problem)

Blending example:
 
 

Feed

Calories per lb

Mineral units per lb

Vitamin units per lb

Cost per lb

Oat

100

200

200

5 cents

Corn

100

400

100

3 cents

Min. Requirement

4000

10000

5000

 

Question: How many pounds of oat and corn should be fed to each cattle per day to minimize feed cost?

Formulation:

Objective: minimize feed cost

Constraints: calories, mineral, and vitamin requirements

Decision variables:
     Let X1 be the pounds of oat to be used
     Let X2 be the pounds of corn to be used

Minimize C = 5X1 + 3 X2

s.t.

100X1 + 100X2 >= 4000 [calorie requirement]

200X1 + 400X2 >= 10000 [mineral requirement]

200X1 + 100X2 >= 5000 [vitamin requirement]

X1, X2 >= 0 [non-negativity constraint]

Product Mix example:
 
 

 

BeltA

BeltB

Availability

Buckle A

1

 

400

Buckle B

 

1

700

Leather (yd)

1

1

800

Labor (hr)

1

2

1000

Profit ($)

40

30

 

Question: How many belts of each type should the company make per day in order to maximize profit?

Formulation:

Objective: Maximize profit

Constraints: resource limitation (buckle, leather, and labor)

Decision variables:
     Let X1 be the number of type A belts to be made per day
     Let X2 be the number of type B belts to be made per day

Maximize Z = 40X1 + 30X2

s.t.

X1 <= 400 [buckle A availability]

X2 <= 700 [buckle B availability]

X1 + X2 <= 800 [leather availability]

X1 + 2X2 <= 1000 [labor availability]

X1, X2 >= 0 [non-negativity constraint]

Exercise:
pg.702, problems 4a, 5a, 6a, 7

Graphical solution method
1.  Assign X1 and X2 to the vertical and horizontal axes
2.  Plot constraint equations
3.  Determine the area of feasibility
4.  Plot the objective function
5.  Find the optimum point

Corner point solution method
1.  Find the co-ordinates of each corner point (by simultaneously solving the equations of a pair of intersecting lines)
2.  Substitute the values of the co-ordinates in the objective function

Exercise:
Pg.701, problems 1, 2, 4b, 5b, 6b