2.

Sewing
Liner Programming 3. Finishing (inserting umbrella holder, club
Unit 6.1 separators, etc.
4. Inspection and packaging

This production information is summarized below:

Linear Programming
Production Time (hours)
 Involves choosing a course of action when the Department Standard Deluxe
mathematical model of the problem contains only Bag Bag
linear functions Cutting and Dyeing 7/10 1
 A problem-solving approach developed to help Sewing 1/2 5/6
Finishing 1 2/3
managers make decisions
Inspection and
Has nothing to do with computer programming Packaging 1/10 1/4

The use of the word “programming” here means

“choosing a course of action” Par, Inc., production is constrained by a limited number
of hours available in each department. The director of
- The maximization or minimization of some manufacturing estimates that 630 hours of cutting and
quantity is the objective in all linear programming dyeing, 600 hours of sewing, 708 hours of finishing, and
problems. 135 hours of inspection and packaging will be available
- All linear programming problems have constraints for the production of golf bags during the next three
that limit the degree to which the objective can be months.
pursued.
- A feasible solution satisfies all the problem’s The accounting department analyzed the production
constraints. data and arrived at prices for both bags that will result in
- An optimal solution is a feasible solution that a profit contribution of $10 for every standard bag and $9
results in the largest possible objective function for every deluxe bag produced.
value when maximizing (or small when minimizing) The complete model for the Par, Inc., problem:
- A graphical solution method can be used to solve
a linear program with two variables.

Guidelines for Model Formulation

Problem formulation or modeling – the process of

translating a verbal statement of a problem into a
mathematical statement

o Understand the problem thoroughly

o Describe the objective
o Describe each constraint Graphical Solution Procedure
o Define the decision variables Inequality representing the cutting and dyeing constraint
o Write the objective in terms of the decision 7
variables is: S +1 D ≤ 630
D
o Write the constraints in terms of the decision
variables

A Simple Maximization Problem

Par, Inc., is a small manufacturer of golf equipment and

supplies whose management has decided to move into
the market or medium– and high-priced golf bags. Par,
Inc., produces over the next three months.

Each golf bag produced will require the following

operation:

1. Cutting and dyeing the material

 The optimal solution point is at the intersection of the
Solution points satisfying each of the other three cutting and dyeing and the finishing constraint lines.
constraints

The optimal values of the decision variables S and D

must satisfy dyeing and the finishing constraints
simultaneously

7
S +1 D=630 Dyeing Constraint
D
2
1 S + D≤ 630 Finishing Constraint
3
The exact location of the optimal solution point is
S=540 and D=252. The optimal production quantities
for Par, Inc., are 540 standard bags and 252 deluxe
bags, with a resulting contribution of
10 ( 540 ) +9 ( 252 )=$ 7,668.
Summary of the Graphical Solution Procedure for
Maximization Problems

1. Prepare a graph of the feasible solutions for each of

the constraints
2. Determine the feasible region that satisfies all the
constraints simultaneously
3. Draw an objective function line
4. Move parallel objective function lines toward larger
objective function values without entirely leaving
Identifies the feasible region
the feasible region
5. Any feasible solution on the objective function line 7 + 1D + 1 S1 = 630
with the largest value is an optimal solution
S
10
Slack and Surplus Variables 1 + 5 + + 1S = 600
S D 2
2 6
A linear program in which all the variables are non- 1S + 2 + + 1S = 708
negative and all the constraints are equalities is said to D 3
3
be in standard form
1 + 1 + + 1S = 135
S D 4
Standard form is attained by adding slack variables to 10 4
“less than or equal to” constraints, and by subtracting S , D , S1 , S 2 , S3 , S 4 ≥0
surplus variables from “greater than or equal to”
constraints

Slack and surplus variables represent the difference

between the left and right sides of the constraints

Slack and surplus variables have objective function

coefficients equal to 0.

The complete solution tells management that the Referring to the standard form of the Par, Inc., problem,
production of 540 standard bags and 252 deluxe bags we see that at the optimal solution (S = 540 and D =
will require all available cutting and dyeing time (630 252), the values for the slack variables are
hours) and all available finishing time (708 hours), while
600−480=120 hours of sewing time and
135−117=18 hours of inspection and packaging time
will remain unused. The 120 hours of unused sewing
time and 18 hours of unused inspection and packaging
time are referred to as slack for the two departments.

Constraint Hours required for Hours Unused

S = 540 and D = 252 Available Hours On the other hand, the sewing and the inspection and
packaging constraints are not binding the feasible region
at the optimal solution, which means we can expect
Cutting and 7/10(540) + 1(252) 630 0 some unused time or slack for these two operations.
Dyeing = 630
Extreme Points and the Optimal Solution
Sewing 1/2(540) + 5/6(252) 600 120
= 480 Extreme Points – the corners or vertices of the feasible
Finishing 1(540) + 2/3(252) 708 0 region
= 708
Inspection 1/10(540) + 1/4(252) 1/4 18 An optimal solution to an LP problem can be found at
and = 117
Packaging an extreme point of the feasible region.

When looking for the optimal solution, you do not have

Slack Variables to evaluate all feasible solution points.

 Often slack variables, are added to the formulation You have to consider only the extreme points of the
of a linear programming problem to represent the feasible region.
slack, or idle capacity 5 extreme points of the feasible region or the Par, Inc.,
Problem
Unused capacity makes no contribution to profit; thus,
slack variables have coefficients of zero in the
objective function. After the addition of four slack
variables, denoted as S1 , S 2 , S 3 , and S4 , the
mathematical model of the Par, Inc., problem becomes
Max 10S + 9D + 0 S1 + 0 S2 + 0 S3 + 0 S4
s.t.
 M&D Chemicals produced two products that are sold as
raw materials to companies manufacturing bath soaps
and laundry detergents.

o M&D’s management specified that the combined

production for products A and B must total at least
350 gallons.
o A customer ordered 125 gallons of product A.
o Product A requires 2 hours of processing time per
gallon.
o Product B requires 1 hour of processing time per
gallon.
o 600 hours of processing time are available.
o M&D/s objective is to satisfy these requirements at a
Computer Solutions
minimum total production cost.
LP problems involving 1000s of variables and 1000s of o Production costs are $2 per gallon for product A and
constraints are now routinely solved with computer $3 per gallon for product B.
packages.
After adding the nonnegativity constraints ( A , B ,≥ 0),
Linear programming solvers are now part of many we arrive at the following linear program for the M&D
spreadsheet packages, such as Microsoft Excel. Chemicals problem:
Leading commercial packages include CPLEX, LINGO, Min 2A + 3B
MOSEK, Xpress-MP, and Premium Solver or Excel. s.t.
1A ≥ 125 Demand for product A
1 A + 1B ≥ 350 Total production
2 A + 1B ≤ 600 Processing time
Computer solution to the Par, Inc., problem
A,B≥0

Feasible region for the M&D Chemicals problem

1. Prepare a graph of the feasible solutions for each of

the constraints.
2. Determine the feasible region that satisfies all the
constraints simultaneously.
3. Draw an objective function line.
4. Move parallel objective function lines toward Note that the objective function 2 A +3 B=800
smaller objective function values without entirely intersects the feasible region at the extreme point A =
leaving the feasible region 250, B = 100
5. Any feasible solution on the objective function line The optimal solution to the M&D Chemicals problem
with the smallest value is an optimal solution. shows that the desired total production of
A Simple Minimization Problem A+ B=350 gallons is achieved by using all processing
time: 2 A +1 B=2 ( 250 )+1 ( 100 ) =600 hours
Note that the constraint requiring that product A demand o Is infeasible
be met has been satisfied with A = 250 gallons. In fact, o Has a unique optimal solution
the production of product A exceeds its minimum level o Has alternative optimal solutions
by 250−125=125 gallons o Has an objective function than can be increased
without bound
This excess production for product A is referred to as
surplus A feasible region may be unbounded and yet there may
be optimal solutions. This is common in minimization
Including two surplus variables, S1and S2, for the ≥
problems and is possible in maximization problems.
constraints and one slack variable, S3, for the ≤
constraint, the linear programming model of the M&D Special Cases
Chemicals problem becomes Alternative Optimal Solutions
Max 2A + 3B + 0 S1 + 0 S2 + 0 S3 In the graphical method, if the objective function line is
s.t. parallel to a boundary constraint in the direction of
1A - 1 S1 = 125 optimization, there are alternate optimal solutions,
1A + 1B - 1 S2 = 350 with all points on this line segment being optimal.
2A + 1B + 1 S3 = 600
Let’s return to the Par, Inc., problem. However, now
A , B , S1 , S 2 , S 3 ≥0 assume that the profit for the standard golf bag (S) has
All the constraints are now equalities. decreased to $6.30. The revised objective function
becomes 6.3 S +9 D .
At the optimal solution of A = 250 and B = 100, the
values of the surplus and slack variables are as follows:

Computer Solution
The objective function values at these two extreme
points are identical:

6.3 S +9 D=6.3 ( 300 )+ 9 ( 420 )=5670 and

6.3 S +9 D=6.3 ( 540 ) + 9 ( 252 ) =5670

Furthermore, any point on the line connecting the two
optimal extreme points also provides an optimal solution.

For example, the solution point (S=420 , D=336),

which is halfway between the two extreme points, also
provides the optimal objective function value of
Feasible Region
6.3 S +9 D=6.3(420)+9( 336)=5670
The feasible region for a two-variable LP problem can
be nonexistent, a single point, a line, a polygon, or A linear programming problem with alternative optimal
an unbounded area. solutions is generally a good situation for the manager or
decision maker. It means that several combinations of
Any linear program falls in one of four categories:
the decision variables are optimal and that the manager A disadvantage of using general notation for decision
can select the most desirable optimal solution. variables is that we are no longer able to easily identify
what the decision variables actually represent in the
Infeasibility
mathematical model.
No solution to the LP problem satisfies all the
The advantage of general notation is that formulating a
constraints, including the non-negativity conditions.
mathematical model for a problem that involves a large
Graphically this means, a feasible region does not number of decision variables is much easier.
exist.

Causes include:

o A formation error has been made

o Management’s expectations are too high
o Too many restrictions have been placed on the
problem (i.e. the problem is over-constrained)

Unbounded

The solution to a maximization LP problem is unbounded

if the value of the solution may be made indefinitely large
without violating any of the constraints.

For real problems, this is the result of improper

formulation. (Quite likely, a constraint has been
inadvertently omitted.)

General Linear Programming Notation

We selected decision-variable names of S and D in the

Par, Inc., problem and A and B in the M&D Chemicals
problem to make it easier to recall what these decision
variables represented in the problem.

Although this approach works well for linear programs

involving a small number of decision variables, it can
become difficult when dealing with problems involving a
large number of decision variables.

A more general notation that is often used for linear

programs uses the letter x with a subscript.

In the Par, Inc., problem, we could have defined the

decision variables:

X 1 = number of standard bags

X 2 = number of deluxe bags

In the M&D Chemicals problem, the same variable
names would be used, but their definitions would
change:

X 1 = number of gallons of product A

X 2 = number of gallons of product B