BRE4281-Construction Engineering

Management

Linear Programming and Applications

Dr. Yongtao TAN and Dr. AnnYU

Department of Building & Real Estate
The Hong Kong Polytechnic University

1
 Lecture Outline

 Introduction
 Defining linear programming
 Procedures of linear programming
 Formulating linear programming problems
 Case study
 In-class exercises

2
 Introduction

 Many companies and research programs are integrating optimization

technology into their day-to-day operations. Manipulating models
and optimization software is becoming an increasingly popular skill in
various contexts.

o Linear Programming
o Dynamic Programming

George B. Dantzig

3
 Introduction
Decision to be made???

 A furniture manufacturer makes high quality wooden tables and chairs.

Its profit is $15 per chair and $21 per table.
 Weekly production is constrained by available labor and wood. Each
chair requires 4 labor hours and 8 board feet of wood while each table
requires 3 labor hours and 12 board feet of wood. Available wood is
2400 board feet and available labor is 920 hours.
 Management also requires at least 40 tables and at least 4 chairs be
produced for every table produced.
 To maximize profits, how many chairs and tables should be produced?

4
Introduction

 Make most effective use of an organisation’s resources,

which include:
o machinery
o labour
o money
o Time
o Space
o raw materials

 Linear programming is a mathematical technique to help

managers in planning and decision making to resource allocation

5
Defining linear programming

o Occasionally do with computer programming

o Refers to modelling and solving a problem mathematically
o A mathematical technique to help management decide how to make
most effective use of an organisation’s resources

6
 Defining linear programming

 Requirements of a linear programming problem

o All problems seek to achieve the objectives of optimization: either to maximise

or minimise
o The presence of restrictions or constraints, limit the degree to which we can
pursue our objectives
o Alternatives to choose from the objective and constraints must be expressed in
terms of linear equations or inequalities

7
Defining linear programming

 Basic assumptions of linear programming

o Conditions of certainty exist

o Proportionality exists in the objective and constraints
o Additivity
o Divisibility
o All answers or variables are non-negative

8
 Procedures of linear programming

 1. Define the problem

o What is the background of the problem concerned, objective (which may

be about maximizing the profits with given resources or conditions, or
minimizing the costs)

9
 Procedures of linear programming

 2. Analyse the given data in order to develop a mathematical model

to represent the real situation.

o a. Define unknowns and variables;

o b. Define objective and formulate the algebraic expression for the

objective

o c. Define the conditions or constraints and formulate the algebraic

expressions for the constraints and conditions.

10
 Procedures of linear programming

 3. Find out the optimum solution

- techniques available: graphic method, simplex method, computing

packages, etc.

 4. Interpret the solution results

- Implication, sensitivity analysis

11
 Procedures of linear programming

 Standard Form of Linear Programming

Why it's difficult:

 Lots of variables (n of them)

 Lots of boundaries to check (the inequalities)

12
Procedures of linear programming

 Graphic method

 Procedures

o Formulate the feasible area (to meet all the constraint algebraic
expressions) in a coordinate diagram
o Define objective lines
o Identify the maximum objective value
o Sensitivity analysis

13
 Formulating linear programming problems

o One of the most common linear programming applications is the product mix
problem
o Two or more products are usually produced using limited resources such as
personnel, machines, raw materials, etc.
o The profit that the firm seeks to maximise is based on the profit contribution per
unit of each product
o The company would like to determine how many units of each product it should
produce so as to maximise overall profit given its limited resources

https://www.youtube.com/watch?v=YPn4yHM1YsU
14
 Formulating linear programming problems

 Case study:
o The Wyndor Glass Co. produces high-quality glass products, including windows and
glass doors. It has three plants. Aluminum frames and hardware are made in Plant A,
wood frames are made in Plant B, and Plant C cuts the glass and assembles the products.
Because of declining earnings, manager decides that unprofitable products are to be
discontinued, and product capacity will be reassigned to launch two new products. The
two new products include:

1. An 8-foot glass door with aluminum framing

2. A 4×6 foot double-hung wood-framed window

The product 1 requires production capacity in Plants A and C only. The product 2 needs
only Plants B and C. The marketing department figures that the company can sell
15 as many of each product as they can produce.
 Formulating linear programming problems

o To produce one batch of product 1, the workers in plant A are required to work for 1 hour
and workers in plant C are required to work for 3 hours per week. To produce one batch
of product 2, the workers in plant B are required to work for 2 hours and workers in plant
C are required to work for 2 hours per week.

o Available working hours per week in each plant for the new products is 4 hours, 12 hours,
and 18 hours respectively.

o The profit of product 1 and 2 is $3,000 and $5,000 respectively.

Question:

It is not clear how many of the two products should be produced

respectively per week that would be most profitable.
16
 Formulating linear programming problems

Determine the Problem

o Determine what the production rates should be for the two products in order to
maximize their total profit, subject to restrictions imposed by the limited production
capacities available in the three plants.

o Each product will be produced in batches, so the production rate to be determined is the
number of batches per week. Any combination of production rates that satisfies these
restrictions is permitted, including producing none of one product and as much as
possible of the other.

17
 Formulating linear programming problems

Collect Relevant Data

1. Number of hours of production time available per week in each plant for the
new products.
2. Number of hours of production time used in each plant for each batch of each
new product.
3. Profit per batch produced of each new product.

Plant A Plant B Plant C Profit / batch

Product 1 1 -- 3 $3,000
Product 2 -- 2 2 $5,000
Available working
4 12 18
hours / week
18
Formulating linear programming problems

Formulate the Mathematical Model

We recognize this as a linear programming problem of the classic product mix type.

o X1=number of batches of product 1 produced per week

o X2=number of batches of product 2 produced per week
o Z=total profit per week (thousands of dollars) from producing the two products

Thus, X1 and X2 are decision variables and Z is the objective function (maximized). Based on
collected parameters, we have:

Z = 3 X1+ 5 X2

19
 Formulating linear programming problems

Describe Constraints

The data indicates that each batch of product 1 produced per week uses 1 hour of
production time per week in Plant A, whereas only 4 hours per week are available in Plant
A. The restriction is written mathematically as
X1 ≤ 4
Similarly, Plant B imposes the restriction

2X2 ≤ 12
The number of hours of production time used per week in Plant C given X1 and X2 is 3
X1+ 2X2 (3 hours of production per batch of product 1 and 2 hours per batch of product 2).
The available capacity in Plant C provides the constraint:
3X1 + 2X2 ≤ 18
Finally, production rates cannot be negative, so it is necessary to impose restrictions
20 X1 ≥ 0 and X2 ≥ 0
 Formulating linear programming problems

Algebraic Formulation

Max Z = 3X1 + 5X2

subject to the restrictions

X1 ≤ 4

2X2 ≤ 12
3 X1 +2X2 ≤ 18
and

X1≥ 0; X2≥ 0

21
 Formulating linear programming problems

Graphical Solution: the Feasible Region

This problem has only two dimensions,

so a graphical procedure can be
employed. We use label the axes as X1
and X2.

The first step is to identify on the

graph values of (X1; X2) which are
feasible (consistent with the
restrictions).

22
 Formulating linear programming problems

Finding the optimum solution - corner point solution method

o It involves looking at the profit at every corner point of the feasible region
o An optimum solution to any problem will lie at a corner point, or extreme point, of
the feasible region
o Hence, it is only necessary to find the values of the variables at each corner
o The maximum profit or optimum solution will lie at one (or more) of them

23
 Formulating linear programming problems

Finding the optimum solution

Note that the objective
function Z = 3 X1 + 5 X2
can be rewritten in slope-
intercept form.

The five corner points /

extreme points:
(0, 0) (Z=0)
(4, 0) (Z=12)
(0, 6) (Z=30)
(2, 6) (Z=36)
24
(4, 3) (Z=27)
Formulating linear programming problems

Solution

o The optimal solution is the point (2,6) where X1 =2; X2 = 6; with Z=36. This implies 2 batches
of product 1 and 6 batches of product 2 will be produced per week, providing a total profit of
$36,000 per week.

o Next step: assess practical realities and possible reassess the model. Subject the model to
both parametric and structural sensitivity analysis.

25
Formulating linear programming problems

 Application

What if the profit of product 1 and product 2 changed to $5000 and $3000
respectively?

Corner/extreme points Z values

(0, 0) 0
(4, 0) 20
(0, 6) 18
(2, 6) 28
(4, 3) 29

26
Formulating linear programming problems

 Application

Z=29=5x1+3x2

27
 Thank You !

28