AAOC C222: OPTIMISATION

Text Book:
Operations Research: An Introduction By Hamdy A.Taha (Pearson Education) th 7 Edition

Reference Books:
1. Hadley, G: Linear Programming, Addison Wesley 2. Pant, J.C: Optimization, Jain Brothers

3. Hillier & Lieberman Introduction to Operations Research, Tata McGraw-Hill

4. Bazaraa, Jarvis & Sherali Linear Programming and Network Flows, John Wiley

You may view my lecture slides in the following site.

http://discovery.bitspilani.ac.in/discipline/math/msr/index. html

The formal activities of Operations Research (OR) were initiated in England during World War II when a team of British scientists set out to make decisions regarding the best utilization of war material. Following the end of the war, the ideas advanced in military operations were adapted to improve efficiency and productivity in the civilian sector. Today, OR is a dominant and indispensable decision making tool.

Example: The Burroughs garment company manufactures men's shirts and womens blouses for Walmark Discount stores. Walmark will accept all the production supplied by Burroughs. The production process includes cutting, sewing and packaging. Burroughs employs 25 workers in the cutting department, 35 in the sewing department and 5 in the

packaging department. The factory works one 8-hour shift, 5 days a week. The following table gives the time requirements and the profits per unit for the two garments:

Minutes per unit

Garment Cutting Sewing Packaging Unit profit($)

Shirts Blouses

20 60

70 60

12 4

8.00 12.00

Determine the optimal weekly production schedule for Burroughs.

Solution: Assume that Burroughs produces x1 shirts and x2 blouses per week. Profit got = 8 x1 + 12 x2 Time spent on cutting = 20 x1 + 60 x2 mts Time spent on sewing = 70 x1 + 60 x2 mts Time spent on packaging =12 x1 + 4 x2 mts

The objective is to find x1, x2 so as to

maximize the profit z = 8 x1 + 12 x2

satisfying the constraints:

20 x1 + 60 x2 25 40 60

70 x1 + 60 x2 35 40 60
12 x1 + 4 x2 5 40 60 x1, x2 0, integers

This is a typical optimization problem.

Any values of x1, x2 that satisfy all the constraints of the model is called a feasible solution. We are interested in finding the optimum feasible solution that gives the maximum profit while satisfying all the constraints.

More generally, an optimization problem looks as follows: Determine the decision variables x1, x2, , xn so as to optimize an objective function f (x1, x2, , xn) satisfying the constraints gi (x1, x2, , xn) bi (i=1, 2, , m).

Linear Programming Problems(LPP)

An optimization problem is called a Linear Programming Problem (LPP) when the objective function and all the constraints are linear functions of the decision variables, x1, x2, , xn. We also include the non-negativity restrictions, namely xj 0 for all j=1, 2, , n. Thus a typical LPP is of the form:

Optimize (i.e. Maximize or Minimize) z = c1 x1 + c2 x2+ + cn xn subject to the constraints: a11 x1 + a12 x2 + + a1n xn b1 a21 x1 + a22 x2 + + a2n xn b2 . . . am1 x1 + am2 x2 + + amn xn bm x1, x2, , xn 0

A LPP satisfies the two properties:

Proportionality and additivity Proportionality means the contributions of each decision variable in the objective function and its requirements in the constraints are directly proportional to the value of the variable.

Additivity stipulates that the total contributions of all the variables in the objective function and their requirements in the constraints are the direct sum of the individual contributions or requirements of each variable.

 We shall first look at formulation of some LPPs,

Graphically solve some LPPs involving two decision variables

Study some mathematical preliminaries regarding the solutions of LPPs Finally look at the Simplex method of solving a LPP

Wild West produces two types of cowboy hats. Type I hat requires twice as much labor as a Type II. If all the available labor time is dedicated to Type II alone, the company can produce a total of 400 Type II hats a day. The respective market limits for the two types of hats are 150 and 200 hats per day. The profit is $8 per Type I hat and $5 per Type II hat. Formulate the problem as an LPP so as to maximize the profit.

Solution: Assume that Wild West produces x1 Type I hats and x2 Type II hats per day. Per day Profit got = 8 x1 + 5 x2

Assume the time spent in producing one type II hat is c minutes. Labour Time spent is (2 x1 + x2) c minutes

The objective is to find x1, x2 so as to

maximise the profit z = 8 x1 + 5 x2

satisfying the constraints:

(2 x1 + x2 ) c 400 c
x1
x2

150 200

x1, x2 0, integers

That is: The objective is to find x1, x2 so as to maximise the profit z = 8 x1 + 5 x2 satisfying the constraints: 2 x1 + x2 400 x1
x2

150 200

x1, x2 0, integers

Feed Mix problem: The manager of a milk diary decides that each cow should get at least 15, 20 and 24 units of nutrients A, B and C respectively. Two varieties of feed are available. In feed of variety 1(variety 2) the contents of the nutrients A, B and C are respectively 1(3), 2(2), 3(2) units per kg. The costs of varieties 1 and 2 are respectively Rs. 2 and Rs. 3 per kg. How much of feed of each variety should be purchased to feed a cow daily so that the expenditure is least?

Trim Loss problem: A company has to manufacture the circular tops of cans. Two sizes, one of diameter 10 cm and the other of diameter 20 cm are required. They are to be cut from metal sheets of dimensions 20 cm by 50 cm. The requirement of smaller size is 20,000 and of larger size is 15,000. The problem is : how to cut the tops from the metal sheets so that the number of sheets used is a minimum. Formulate the problem as a LPP.

A sheet can be cut into one of the following three patterns: 10 10

Pattern I Pattern II 20 10 20 Pattern III 20

10
10

Pattern I: cut into 10 pieces of size 10 by 10 so as to make 10 tops of size 1 Pattern II: cut into 2 pieces of size 20 by 20 and 2 pieces of size 10 by 10 so as to make
2 tops of size 2 and 2 tops of size 1 Pattern III: cut into 1 piece of size 20 by 20 and 6 pieces of size 10 by 10 so as to make 1 top of size 2 and 6 tops of size 1

So assume that x1 sheets are cut according to pattern I, x2 according to pattern II, x3 according to pattern III

The problem is to
Minimize z = x1 + x2 + x3

Subject to 10 x1 + 2 x2 + 6 x3 20,000
2 x2 + x3 15,000

x1, x2, x3 0, integers

A Post Office requires different number of full-time employees on different days of the week. The number of employees required on each day is given in the table below. Union rules say that each full-time employee must receive two days off after working for five consecutive days. The Post Office wants to meet its requirements using only full-time employees. Formulate the above problem as a LPP so as to minimize the number of fulltime employees hired.

Requirements of full-time employees day-wise

Day 1 - Monday 2 - Tuesday 3 - Wednesday 4 - Thursday 5 - Friday 6 - Saturday 7 - Sunday No. of full-time employees required 10 6 8 12 7 9 4

Solution: Let xi be the number of full-time employees employed at the beginning of day i (i = 1, 2, , 7). Thus our problem is to find xi so as to

Minimize z x1 x2 x3 x4 x5 x6 x7 Subject to x1 x4 x5 x6 x7 10 (Mon)

x1 x2 x5 x6 x7 6 (Tue) x1 x2 x3 x6 x7 8 (Wed) x1 x2 x3 x4 x7 12 (Thu) x1 x2 x3 x4 x5 7 (Fri) x2 x3 x4 x5 x6 9 (Sat) x3 x4 x5 x6 x7 4 (Sun)

xi 0. integers

BITS wants to host a Seminar for five days. For the delegates there is an arrangement of dinner every day. The requirement of napkins during the 5 days is as follows:
Day
Napkins Needed

1
80

2
50

3
100

4
80

5
150

Institute does not have any napkins in the beginning. After 5 days, the Institute has no more use of napkins. A new napkin costs Rs. 2.00. The washing charges for a used one are Rs. 0.50. A napkin given for washing after dinner is returned the third day before dinner. The Institute decides to accumulate the used napkins and send them for washing just in time to be used when they return. How shall the Institute meet the requirements so that the total cost is minimized ? Formulate as a LPP.

Solution Let xj be the number of napkins purchased on day j, j=1,2,..,5 Let yj be the number of napkins given for washing after dinner on day j, j=1,2,3 Thus we must have x1 = 80, x2 = 50, x3 + y1 = 100, x4 + y2 = 80 x5 + y3 = 150 Also we have y1 80, y2 (80 y1) + 50 y3 (80 y1) + (50 y2) + 100

Thus we have to Minimize

z = 2(x1+x2+x3+x4+x5)+0.5(y1+y2+y3)
Subject to

x1 = 80, x2 = 50, x3 +y1 =100,

x4 + y2 = 80, x5 + y3 = 150, y1 80, y1+y2 130, y1+y2+y3 230,

all variables 0, integers

There are many Software packages available to solve LPP and related problems.

Your book contains a CD having the package TORA probably developed by the author.
There is also Microsofts Excel Solver. Normally this would not have been loaded; you mut check whether it is loaded. There is also a commercial package LINGO

 Dr. J C Pants book contains in the end a C code for solving some of the LPP problems (of course developed by some of your seniors).

You may yourself develop programs to solve LPP problems.