INTEGER PROGRAMMING

Rennie Boy L. Lanutan, LPT

INTRODUCTION

 In the linear programming models formulated and solved in the

previous chapters, the implicit assumptions was that solutions
could be fractional or real numbers (ex. Non-integer).
 However, non-integer solutions are not always practical.
 When only integer solutions are practical or logical, it is
sometimes assumed that non-integer solution values can be
“round off” to be the nearest feasible integer values.
 This method would cause little concern if, for example, x1 =
8,000.4 nails were rounded off to 8,000 nails because nails cost
only a few cents a piece.
INTRODUCTION

 However, if we are considering the production of jet

aircraft and x1=7.4 jet airliners, rounding off could affect
profit (or cost) by millions or dollars.
 In this case we need to solve the problem so that an
optimal integer solution is guaranteed.
 In this chapter the different forms of integer linear
programming models are presented
INTEGER PROGRAMMING MODELS

 There are three basic types of integer linear programming models-a

total integer model, a 0-1 integer model, and a mixed integer model
1. In a total integer model, all the decision variables are required to
have integer solution values.
2. in a 0-1 integer model, all the decision variables have integer
values of zero or one.
3. Finally, in a mixed integer model, some of the decision variables
(but not all) are required to have integer solutions.
The following three examples demonstrate these types of integer
programming models.
1. A TOTAL INTEGER MODEL EXAMPLE

The owner of a machine shop is planning to expand by purchasing

some new machines-presses and lathes. The owner has estimated
that each press purchased will increase profit by $100 per day and
each lathe will increase profit by $150 daily. The number of
machines the owner can purchase is limited by the cost of the
machines and the available floor space in the shop. The machine
purchase prices and space requirements are as follows:
1. A TOTAL INTEGER MODEL EXAMPLE

The owner has a budget of $4o,000 for purchasing machines and

200 square feet of available floor space. The owner wants to know
how many of each type of machine to purchase to maximize the
daily increase in profit.
 The linear programming model for an integer programming
problem is formulated in exactly the same way as the linear
programming examples in chapter 2. The only difference is that
in this problem, the decision variables are restricted to integer
values because the owner cannot purchase a fraction, or portion
of a machine.
1. A TOTAL INTEGER MODEL EXAMPLE

The linear programming model follows:

maximize Z=$100x1 + $150x2
subject to
$8,000x1+$4,000x2
15x1 + 30x2 200 ft.2
x1, x2
where
x1=number of presses
x2=number of lathes
1. A TOTAL INTEGER MODEL EXAMPLE

The decision variables in this model are restricted

to whole machines. The fact that both decision
variables, x1 and x2 can assume any integer value
greater than or equal to zero is what gives this
model its designation as a total integer model.
2. A 0-1 INTEGER MODEL EXAMPLES

A community council must decide which recreation facilities to construct in

its community. Four new recreation facilities have been proposed-a
swimming pool, a tennis center, an athletic field and a gymnasium. The
council wants to construct facilities that will maximize the expected daily
usage by the resident of the community, subject to land and cost limitations.
The expected daily usage and cost and land requirements for each facility
follows:
2. A 0-1 INTEGER MODEL EXAMPLES

The community has a $120,000 construction budget and

12 acres of land. Because the swimming pool and tennis
center must be built on the same part of the land parcel,
however, only one of these two facilities can be
constructed. The council wants to know which of the
recreation facilities to construct to maximize the
expected daily usage.
2. A 0-1 INTEGER MODEL EXAMPLES

The model for this problem is formulated as follows:

maximize Z = 300x1 + 90x2 + 400x3 + 150x4
subject to
$35,000x1 + 10,000x2 + 25,000x3 + 90,000x4
4x1 + 2x2 + 7x3 + 3x4 12 acres
x1 + x2 1 facility
where
x1, x2, x3, x4= 0 or 1
2. A 0-1 INTEGER MODEL EXAMPLES

where
x1 = construction of a swimming pool
x2 = construction of a tennis center
x3 = constructions of an athletic field
x4 = construction of a gymnasium
 In this model, the decision variables can have a solution value of either zero
or one. If a facility is not selected for construction, the decision variable
representing it will have the value of zero. If a facility is selected, its
decision variable will have a value of one.
2. A 0-1 INTEGER MODEL EXAMPLES

 The last constraint, x1 + x2 1, reflects the contingency that either the

swimming pool (x1) or the tennis center (x2) can be constructed, but not
both. In order for the sum of x1 and x2 to be less than or equal to one, either
of the variable can have a value of one, or both variables can equal zero.
This is also refer to as Mutually Exclusive Constraints.
 If the community had specified that either the swimming pool (x1) or the
tennis center (x2) must be built, but not both, then the last constraint would
become an equation, x1 + x2 = 1. This would result in a solution that would
include x1 = 1 or x2 = 1, but both would not equal one (nor would both equal
zero). In this manner, the model forces a choice between the two facilities.
For this reason, it is often called a Multiple Choice Constraint.
2. A 0-1 INTEGER MODEL EXAMPLES

 A variation of the multiple choice constraint can be used to

formulate a situation in which some specific number of facilities
out of the total must be constructed. For example, if the
community council had specified that exactly two of the four
facilities must be built, this constraint would be formulated as
x1 + x2 + x3 + x4 = 2
 If, alternatively, the council had specified that no more than two
facilities must be constructed, the constraint would be
x1 + x2 + x3 + x4 2
2. A 0-1 INTEGER MODEL EXAMPLES

 Another type of 0-1 model constraint is a Conditional Constraint. In a

conditional constraint, the construction of one facility is conditional
upon the construction of another. Suppose, for example, that the pet
project of the head of the community council is the swimming pool, and
she also believes the tennis center is frivolous. The council head is very
influential, so the rest of the council knows that the tennis center has
no chance of being selected if the pool is not selected first. However,
even if the pool is selected, there is no guarantee that the tennis center
will also be selected. Thus, the tennis center (x 2) is conditional upon
construction of the swimming pool (x1). This condition is formulated as
x2 x 1
2. A 0-1 INTEGER MODEL EXAMPLES

 Notice that the tennis center (x2) cannot equal one (ex. be selected) unless
the pool (x1) equals one. If the pool (x1) equals zero (ex. it is not selected),
then the tennis center (x2) must also equal zero. However this condition
does allow the pool (x1) to equal one and be selected and the tennis center
(x2) to equal zero and not be selected
 A variation of this type of conditional constraint is the Corequisite
Constraint, wherein if one facility is constructed, the other one will also be
constructed and vice versa. For example, suppose the council work out a
political deal among themselves, wherein if the pool (x1) is accepted, the
tennis center (x2) must also be selected and vice versa. This constraint is
written as
x2 = x1
3. A MIXED INTEGER MODEL EXAMPLE

Nancy Smith has $250,000 to invest in three alternative

investments-condominiums, land, and municipal bonds. She
wants to invest in the alternatives that will result in the greatest
return on investment after 1 year.
Each condominium cost $50,000 and will return a profit of $9,000
if sold at the end of 1 year; each acre of land costs $12,000 and will
return a profit of $1,500 at the end of 1 year ; and each municipal
bond costs $8,000 and will result in a return of $1,000 if sold at the
end of 1 year. In addition, there are only 4 condominiums, 15 acres
of land, and 20 municipal bonds available for purchase.
3. A MIXED INTEGER MODEL EXAMPLE

The linear programming model for this problem is formulated as follows:

maximize Z = $9,000x1 + $1,500x2 +$1,000x3
subject to
$50,000x1 + $12,000x2 + $8,000x3 $250,000
x1 4 condominiums
x2 15 acres
X3 20 bonds
x2 0
x1, x3 0 and integer
3. A MIXED INTEGER MODEL EXAMPLE

where
x1 = condominium purchased
x2 = acres of land purchased
x3 = bonds purchased
 Notice that in this model, the solution values for condominiums (x1) and
municipal bonds (x3) must be integers. It is not possible to invest in a
fraction of a condominium or to purchase part of a bond. However, it is
possible to purchase less than an acre of land (ex. a portion of an acre).
Thus, two of the decision variables (x1 and x2) are restricted to integer
variables, whereas the other variable (x2) can take on any real value greater
than or equal to zero.
INTEGER PROGRAMMING GRAPHICAL
SOLUTION
 It may seem logical that an easy way to solve integer programming
problems is to round off fractional solution values to integer values.
However, that can result in less-than-optimal (ex. suboptimal) results. This
outcome can be seen by using graphical analysis. As an example, consider
the total integer model for the machine shop formulated in the previous
sections:
maximize Z = $100x1 + $150x2 subject to
$8,000x1 + $4,000x2 $40, 000
15x1 + 30x2 200
x1, x2 0 and integer
Integer Programming Graphical Solution
We use Excel to solve this model as a regular linear programming model without
the integer requirements, as shown in Exhibit 5.1 (Because not all of the students
have desktop or laptop computers that they can use to solve this model we will
just be presenting the answers as a references in this presentation).
INTEGER PROGRAMMING GRAPHICAL
SOLUTION
 Notice that this model results in a non-integer solution of x1 = 2.22 presses,
and x2 =5.56 lathes. Because the solution values must be integers, let us first
round off these two values to the closest integer values, which are x1 = 2 and
x2 = 6. However, if we substitute these values into the second model
constraint, we find that this integer solution violates the constraint and thus
is infeasible.
15x1 + 30x2 200
15(2) + 30(6) 200
210 200
INTEGER PROGRAMMING GRAPHICAL
SOLUTION
 In a model in which the constraint are all ( and the
constraint coefficients are positive), a feasible solution
is always ensured by rounding down. Thus, a feasible
integer solution for this problem is
x1 = 2
x2 = 5
Z = $950
INTEGER PROGRAMMING GRAPHICAL
SOLUTION
 However, one of the difficulties, of simply rounding
down non-integer values is that another integer solution
may result in a higher profit (ex. in this problem, there
may be an integer solution that will result in a profit
higher than $950).
 In order to determine whether a better integer solution
exist, let us analyze the graph of this model, which is
shown in figure 5.1.
INTEGER PROGRAMMING GRAPHICAL
SOLUTION
In figure 5.1, the dots indicate integer solution points, and
the point x1 = 2, x2=5 is the rounded down solution.

Notice that as the objective function edge moves

outward through the feasible solution space, there is an
integer solution point that yields a greater profit than
The rounded-down solution. This solution point
is x1= 1, x2=6. At this point, Z=$1,000 which is $50

more profit per day for the machine shop than is

realized by the rounded-down integer solution.
This solution point is x1 = 1, x2 = 6. At this point, Z=$1000

which is $50 more profit per day for the machine

shop than is realized by the rounded-down integer solution.
INTEGER PROGRAMMING GRAPHICAL
SOLUTION
 This graphical analysis explicitly demonstrates the error
that can result from solving an integer programming
problem by simply rounding down. In the machine shop
example, the optimal integer solution is x1=1, x2=6
instead of the rounded-down solution, x1=2, x2=5 (which
is often called the suboptimal solution or result).
Because erroneous results are cause by rounding down
regular solution, a more direct approach for solving
integer problems is required.
INTEGER PROGRAMMING GRAPHICAL
SOLUTION
 The traditional approach for solving integer programming problems is the branch
and bound method. It is a mathematical solution approach that can be applied to a
number of different types of problems.
 The branch and bound method is based on the principle that the total set of
feasible solutions (such as the feasible area in figure 5.1 ) can be partitioned into
smaller subsets of solution.
 These smaller subsets can than be evaluated systematically until the best solution
is found. The branch and bound method is a tedious and often complex
mathematical processes.
 Fortunately, both Excel and QM for windows have the capability to solve integer
programming problems, and thus we will rely on the computer to solve the
different types of integer models demonstrated in this chapter.
Solution of the 0-1 Model with Excel
Exhibit 5.6 shows the solution: x1=swimming pool, x2=0 Tennis Center, x3=1 Athletic Field, x4=0
gymnasium, Z=700 people per day expected usage
Solution of the Mixed Integer Model with Excel
Exhibit 5.13 shows the solution: x1=4 condominiums, x2=4.17 acres, x3=0 bonds, Z=$42,250
SUMMARY

 In this chapter, we found that simply rounding off non-integer

simplex solution values for models requiring integer solutions is
not always appropriate. Rounding off can often lead to
suboptimal results. Therefore, direct solutions approaches using
the computer are needed to solve the three forms of linear
integer programming models-total integer models, 0-1 integer
models, and mixed integer models.
 Having analyzed integer problems and the basic knowledge for
solving them, we have now covered most of the basic forms of
linear programming models and solution techniques.