OR2 INTEGER PROGRAMMING MODEL

1. INTRODUCTION
A integer programing problem is an optimization problem for which is done the
following:

1-Attempt to maximize ( or minimize ) a objective function of the decision

variables.

2-The values of the decision variables must satisfy a set of constraints. Each

constraint must be a linear equation or inequality.

3-A sign restriction is associated with each variable.

There are theree basic types of integer linear proramming models ( a total

integer model, a 0-1 integer model, and a mixed integer model ). Integer programing
problems can be solved also by using Excel spreadsheets and QM for Windows.

Integer programing is gaining importance by day by for almost every company.

In this report, the most appropriate solution is tried to be found for Atlantic

Management System.

1.1.Problem Definition
Problem that is tried to be solved in this report is given below in Figure 1.1.
1
2
 Figure 1.1: Problem Definition and Data

Problem given above is an integer programming case which is asked for

number of offices that should open and location of it from the given choices. Then the
optimum allocation of the available employees to the new offies is carried out.

1.2. Solution Methods

Optimum solution is found by using 2 computer programmes :

1- LINDO SOFTWARE PACKAGE

2- MICROSOFT OFFICE EXCEL QM APPLICATION

3
 2. APPLICATION OF SOLUTION METHODS

2.1. LINDO Software Package Soluton

In LINDO, steps below is done to formulate the problem:

1- Variables used in the program is defined clearly

2- Goal of the program is written (Maximization)
3- Constraints are defined
4- Types of the variables are defined

Model written in LINDO is given below Figures.

Figure 1.1. Variables Used in LINDO

4
Figure 1.2. Goal of the Model and Constraints

5
 Figure 1.3. Types of the Variables

2.2. Mcrosoft Offce Excel QM Applcaton

Steps below are applied to solve in Excel QM :

Step1: Required tables are constructed by available data on hand.

Table 2.2.1. An Example of Reguired Data on Excel

Staffing
Prospective Ratin Setup Cost Needs( Employees
Office g ($1.000) )
Y1 1 Atlanta 5 1700,0 9
Y2 1 Boston 4 3600 14
Y3 0 Denver 3 2100 8
Y4 0 Detroit 1 2500 12
Y5 0 Miami 6 3100 11
Y6 1 St.Luis 2 2700 7
Y7 1 Washington DC 7 4100 18

6
 Step2: Goal is chosen as maximum.

Step 3: Constraints are entered by choosing cells into the page in the
Solver button.

Figre 2.2.1. Solver Parameters

Excel QM solution views are seen in Appendix 1 in a detailed way.

7
 3. SOLUTIONS
According to the applications given above , optimum solution for Athlantic

Management Systems is found. Optimum solution which maximizes profit ( taken as

rating in the model) is :

Number of employees which are send to Chicago to St.Luis=7 ;

Chicago to DC=11 ;

Chicago to Atlanta=9 ;

Houston to Boston=14;

Houston to DC=7.

Maximized profit(rating ) = 18 .

In the model, Atlanta, Boston ,St. Lus and Washington Dc are chosen as new offices.

Employees are send from Chicago and Boston. The cost of new offices are totally 13,440(in

million dollars ).

APPENDIX 1 :EXCEL QM SOLUTION

8
 9
Prospective Office Staffing
Rating SetupCost ($1.000
Needs(
) Employees) The cost of transferring an employee(1,000$)
Y1 1 Atlanta 5 1700,0 9 Chicago to Atlanta X11 0 19
Y2 1 Boston 4 3600 14 Chicago to Boston X12 0 32
Y3 0 Denver 3 2100 8 Chicago to Denver X13 0 27
Y4 0 Detroit 1 2500 12 Chicago to Detroit X14 0 14
Y5 0 Miami 6 3100 11 Chicago to Miami X15 0 23
Y6 1 St.Luis 2 2700 7 Chicago to St.Luis X16 7 14
Y7 1 Washington DC 7 4100 18 Chicago to DC X17 11 41
Charlotte to Atlanta X21 9 14
MAX 18 Charlotte to Boston X22 0 47
Charlotte to Denver X23 0 31
ExistingOffice Available Employees Charlotte to Detroit X24 0 28
Chicago 24 Charlotte to Miami X25 0 35
Charlotte 19 Charlotte to St.Luis X26 0 18
Pittsburgh 16 Charlotte to DC X27 0 53
Houston 21 Pittsburgh to Atlanta X31 0 16
Pittsburgh to Boston X32 0 39
CONSTRAINTS Pittsburgh to Denver X33 0 26
S.t Pittsburgh to Detroit X34 0 23
ST1 13440,00 <= 14000 Budget constraint Pittsburgh to Miami X35 0 31
ST2 9 9 Employee constraint for Atlanta Pittsburgh to St.Luis X36 0 19
ST3 14 14 Employee constraint for Boston Pittsburgh to DC X37 0 48
ST4 0 8 Employee Contstraint for Denver Houston to Atlanta X41 0 22
ST5 0 12 Employee Contstraint for Washington , DC Houston to Boston X42 14 26
ST6 0 11 Employee Contstraint for Miami Houston to Denver X43 0 21
ST7 7 7 Employee Contstraint for St. Luis Houston to Detroit X44 0 18
ST8 18 18 Employee Contstraint for Detroid Houston to Miami X45 0 28
ST9 18 <= 24 Available Employees in Chicago Ofce Houston to St.Luis X46 0 24
ST10 9 <= 19 Available Employees in Houston Ofce Houston to DC X47 7 43
ST11 0 <= 16 Available Employees in Pittssburg Ofce
ST12 21 <= 21 Available Employees in Charlotte Ofce
Constraint of at least one new ofce
ST13 1 >= 1 in the midwest ( Detroit and/or St.Luis);
Constraint of one new ofce in the
ST14 1 1 southeast ( Atlanta or Miami);
10