aaoC zC222 optimization

BITS Pilani
K K Birla Goa Campus

Dr. Anil Kumar

Assistant Professor, Department of Mathematics January 5, 2013

Scopes and Objective of the Course

An optimization problem in its simple form is one in which some entity with or without being subjected to certain constraints is minimized or maximized. The entity to be optimized may be profit, cost, time, product, consumer utility, etc. The constraints involved may be manpower, availability of space, raw materials, funds, machine capabilities, governmental controls, etc. There are also optimization problems with more than one objective, such problems will come under Goal Programming.

BITS Pilani, K K Birla Goa Campus

 The subject of optimization is multidisciplinary in nature. Optimization Problems are encountered in physical sciences, engineering, economics, industry, planning and many other areas of human activity. Background needed for undertaking this course is acquaintance with Calculus, Set Theory and Linear Algebra. Objective of the course is to introduce certain standard methods of solving optimization problems.

BITS Pilani, K K Birla Goa Campus

WHO USES OPTIMIZATION?

Effects economic policies on GNP Physical Science Business- plant location, investment policies, marketing budgets Engineering Operations- planning purchases and production Engineering Design- Equipment capacities and flow structures Public Policy- Effects of various policies on well being Econometrics- ces - Statistics (fitting models to data) of - Optimizing systems (minimize free energy)

BITS Pilani, K K Birla Goa Campus

Text Book
T1: H. A. Taha, Operations Research: An Introduction Pearson Education, 8th.edn, 2008

BITS Pilani, K K Birla Goa Campus

Reference Books
R1: C. B. Gupta, Optimization Techniques in Operation Research, I. K. International Publishing House Pvt. Ltd., 2008 . Hillier and Lieberman, Introduction Research, T M H, 8/E, 2006. to Operations

R2:

R3: Igor Griva, Stephen G Nash & Ariela Sofer, Linear & Nonlinear Optimization, Books 24X7 Online Digital Library. R4: J. C. Pant, Introduction to Optimization: Operations Research, Jain Brothers, N. Delhi, 7/E, 2008.

BITS Pilani, K K Birla Goa Campus

Course Description
Linear Programming Dual Problem Transportation Models Assignment Model Advanced Linear Programming Network Models: CPM & PERT Goal Programming Integer Programming Dynamic Programming Game Theory

BITS Pilani, K K Birla Goa Campus

Evaluation Scheme
EC No. Evaluation Component & Type of Examination Quiz Duration Weightage Day, Date, Session, Time

EC-1

Visit website

15%

Visit website Sunday, 10/02/2013 (FN)* 10 AM 12 Noon Sunday, 07/04/2013 (FN)* 9 AM 12 Noon

EC-2

Mid-Semester Test (Closed Book)* Comprehensive Exam (Open Book)*

2 Hours

35%

EC-3

3 Hours

50%

* Please check the details on Taxila website.

BITS Pilani, K K Birla Goa Campus

Contents
Introduction to Mathematical Programming Problems Classification of the problems Linear and Non-Linear programming problems Formulation of Models

BITS Pilani, K K Birla Goa Campus

Convex Sets
An object is convex if for every pair of points within the object, every point on the straight line segment that joins them is also within the object. Line segment between x1 and x2: all points x = x1 + (1-) x2 with 0 1. Convex set: contains line segment between any two points in the set x1, x2 C, 0 1 x1 + (1 )x2 C

BITS Pilani, K K Birla Goa Campus

Example-1

Imagine that you have a 5-week business commitment between Goa (GOI) and Delhi (DEL). You fly out of Goa on Mondays and return on Wednesdays. A regular round-trip ticket costs Rs. 4000, but a 20% discount is granted if the dates of the ticket span a weekend. A one-way ticket in either direction costs 75% of the regular price. How should you buy the tickets for the 5-week period?

BITS Pilani, K K Birla Goa Campus

Example -1

Three question to make decision: What are the decision alternatives? Under what restrictions is the decision made? What is an appropriate objective criterion for evaluating the alternatives?

BITS Pilani, K K Birla Goa Campus

Example -1
Three alternatives: Buy 5 regular GOI-DEL-GOI for departure on Monday and return on Wednesday of the same week. Buy one GOI-DEL, four DEL-GOI-DEL that span weekends, and one DEL-GOI. Buy one GOI-DEL-GOI to cover Monday of the first week and Wednesday of the last week and four DEL-GOI-DEL to cover the remaining legs. All tickets in this alternative span at least one weekend. Objective criterion for evaluating the proposed alternative is the price of the tickets.
BITS Pilani, K K Birla Goa Campus

Example -1
Previous example explains three main components: Alternatives Objective criterion Constraints Situations differ in the details of how each component is developed and constructed.

BITS Pilani, K K Birla Goa Campus

Example-2

To construct a maximum-area rectangle out of a piece of wire of length L inches, What should be the width and height of the rectangle? Remark: Here, the number of alternatives is not finite; namely, the width and height of the rectangle can assume an infinite number of values.

BITS Pilani, K K Birla Goa Campus

Example-2
Let w : width of the rectangle () h : height of the

Based on these definition, the restriction of the situation can be expressed verbally as Width of + height of = half of the length of the wire Width and height can not be negative. Objective of the problem: the maximization of the area of the rectangle.

BITS Pilani, K K Birla Goa Campus

Formulation

BITS Pilani, K K Birla Goa Campus

Terminology
Decision Variables: help in describing the decision to be made. Decision variables may be non negative or unrestricted, depending on the structure of the problem. Usually the decision variables restrictions imposed on them. have non negativity

BITS Pilani, K K Birla Goa Campus

Terminology
Objective function: describes the main objective (or purpose) of the problem. Depending on the nature of the problem, the objective function may be of type maximization or minimization. Constraints: describe the various restrictions imposed on the problem. In any LPP, there are two types of constraints; variable constraints (restrictions on decision variables) and main constraints.

BITS Pilani, K K Birla Goa Campus

Terminology
Three basic steps in constructing a linear programming model: Identify the unknown variables to be determined (decision variables) and represent them in terms of algebraic symbols. Identify all the restrictions or constraints in the problem and express them as linear equations or inequalities which are linear functions of the unknown variables.

BITS Pilani, K K Birla Goa Campus

Terminology
Identify the objective or criterion and represent it as a linear function of the decision variables, which is to be maximized or minimized.

Note: Model building is not a science but primarily an art and comes mainly by practice.

BITS Pilani, K K Birla Goa Campus

Mathematically
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), and xj 0 (j = 1, 2, , n).

BITS Pilani, K K Birla Goa Campus

The Mathematical Programming problem is classified into two classes: 1. Linear Programming Problems (LPP) 2. Nonlinear Programming Problems (NLPP)

BITS Pilani, K K Birla Goa Campus

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.

BITS Pilani, K K Birla Goa Campus

A typical LPP is of the form: Max/Min 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

BITS Pilani, K K Birla Goa Campus

Non-Linear Programming Problems (NLPP)

An optimization problem of the form: Max. / Min. f (X) subject to gi (X) or = or 0, i =1,2,m X 0, where X = (x1, x2, ..xn), is said to be NLPP if either the objective function or the any of the constraints or both are nonlinear function of X.

BITS Pilani, K K Birla Goa Campus

Formulation of a Linear Programming Problem Model

Define the decision variables. Define the objective function. Define the constraints.

BITS Pilani, K K Birla Goa Campus

Example 1
Cycle Trends is introducing two new lightweight bicycle frames, the Deluxe and the Professional, to be made from aluminum and steel alloys. The anticipated unit profits are Rs.10 for the Deluxe and Rs.15 for the Professional. The number of units of each alloy needed per frame is summarized on the next slide. A supplier delivers 100 units of the aluminum alloy and 80 units of the steel alloy weekly. How many Deluxe and Professional frames should Cycle Trends produce each week in order to maximize the profit.

BITS Pilani, K K Birla Goa Campus

Units of each alloy needed per frame

ALUMINUM ALLOY

STEEL ALLOY

DELUXE PROFESSIONAL

2 4

3 2

BITS Pilani, K K Birla Goa Campus

Formulation:
Decision Variables x1 = number of Deluxe frames produced weekly x2 = number of Professional frames produced weekly Objective Maximize total weekly profit Max z = 10x1 + 15x2

BITS Pilani, K K Birla Goa Campus

Constraints Main Constraints: Availability of Aluminum Alloy 2x1 + 4x2 < 100 Availability of Steel Alloy 3x1 + 2x2 < 80 Variable Constraints: x1 > 0, x2 > 0 Non negativity restrictions on the decision variables

BITS Pilani, K K Birla Goa Campus

LPP in Final Form Max z = 10x1 + 15x2 (maximizing the profit) Subject to 2x1 + 4x2 < 100 3x1 + 2x2 < 80 x1 , x2 > 0 ( aluminum constraint) ( steel constraint) (non-negativity constraints)

BITS Pilani, K K Birla Goa Campus

Example 2
A manufacturer produces two types of models M1 and M2. Each M1 model requires 4 hours of grinding and 2 hours of polishing, where as M2 model requires 2 hours of grinding and 5 hours of polishing. The manufacturer has 2 grinders and 3 polishers. Each grinder works for 40 hours a week and each polisher works for 60 hours a week. Profit on M1 is Rs. 3 per unit and profit on M2 is Rs. 4 per unit. whatever is produced in a week is sold in the market. How should the manufacturer allocate his production capacity to two types of models so that the profit is maximum in a week

BITS Pilani, K K Birla Goa Campus

Formulation
Decision Variables: x1 = number of M1 model produced weekly x2 = number of M2 model produced weekly

Objective: Maximize total weekly profit Max z = 3x1 + 4x2

BITS Pilani, K K Birla Goa Campus

Consraints Main constraints For grinding 4x1 + 2x2 2*40 For polishing 2x1 + 5x2 3*60 Variable Constraints: x1 > 0, x2 > 0

BITS Pilani, K K Birla Goa Campus

Final form of the LPP

Max z = 3x1 + 4x2 Subject to 4x1 + 2x2 80 2x1 + 5x2 180 x1 0, x2 0

BITS Pilani, K K Birla Goa Campus

Example-3
A Depo runs buses during the time period 5AM to 1AM. Each bus can operate for 8 hours successively, and then it is directed to workshop for maintenance and fuel. The minimum number of buses required fluctuate with the time intervals. The desired number of buses during different time interval are given in the following table: Time Intervals 5 AM 9 AM 9 AM 1 PM 1 PM 5 PM 5 PM 9 PM 9 PM 1 AM Minimum No. of buses required 5 13 11 14 4
BITS Pilani, K K Birla Goa Campus

It is required to determine the number of buses to operate during different shifts that will meet the minimum requirement while minimizing the total number of daily buses in operation.

BITS Pilani, K K Birla Goa Campus

Formulation Decision Variables xi = number of buses starting at the beginning of the ith period i = 1 to 5. Objective function Min z = x1+x2+x3+x4+x5 Remark : Each bus operates during two consecutive shifts.

BITS Pilani, K K Birla Goa Campus

Constraints x1 + x2 x2 + x3 x3 + x4 x4 + x5 x5 + x1 13 11 14 4 5

Main Constraints

x1 , x2 , x3 , x4 ,x5 0

Variable Constraints

BITS Pilani, K K Birla Goa Campus

LPP in Final Form Min z = x1 + x2+ x3 + x4+ x5 Subject to x1 + x2 13 x2 + x3 11 x3 + x4 14 x4 + x5 4 x5 + x1 5 x1 , x2 , x3 , x4 ,x5 0

BITS Pilani, K K Birla Goa Campus

Example 4
Paper cutting machines are available to cut standard news print rolls into sub rolls. Each standard roll is of 180 cm width and a number of them must be cut to produce smaller sub rolls at the current orders for 30 of width 70 cm, 60 of width 50 cm and 40 of width 30 cm. Formulate the problem so as to minimize the amount of wastes. Ignoring the recycling or other uses for the trim, assume that the length of each required sub roll is the same as that of the standard roll.

BITS Pilani, K K Birla Goa Campus

A standard roll may be cut according to the following patterns.

Number of sub rolls cut on different patterns

Widths ordered in cm 30 50 70 Trim Loss

p1 6 0 0 0

p2 4 1 0

p3 3 0 1

p4 2 1 1 0

p5 2 2 0 20

p6 1 0 2 10

p7 p8 1 3 0 0 0 2 1 10

10 20

BITS Pilani, K K Birla Goa Campus

Formulation
Decision Variables Let xi be the number of the standard news print rolls pieces to cut on the pattern pi, i = 1 to 8.

Objective Function Min z = 10x2 +20x3+ 20x5+ 10x6+ 10x8

BITS Pilani, K K Birla Goa Campus

Constraints 6x1 +4x2+ 3x3+ 2x4+ 2x5 + x6+ x7 = 40 x2 +x4+ 2x5+ 3x7+ 2x8 = 60 x3 +x4+ 2x6+ x7+ x8 = 30 xi 0, i = 1 to 8.

BITS Pilani, K K Birla Goa Campus

Therefore, LPP in Final Form Min z = 10x2 +20x3+ 20x5+ 10x6+ 10x8 Subject to 6x1 +4x2+ 3x3+ 2x4+ 2x5 + x6+ x7 = 40 x2 +x4+ 2x5+ 3x7+ 2x8 = 60 x3 +x4+ 2x6+ x7+ x8 = 30 xi 0, i = 1 to 8.

BITS Pilani, K K Birla Goa Campus

Example 5
A cycle manufacturing company has to meet the following requirement in the coming three months
Month Required Number of cycles October 5,000 November 10,000 December 15,000

The company produces at maximum 15,000 cycles in a month. The production cost of one cycle is Rs 2000. The company does not have a storage place, it hires storage facility. The charges for this are Rs 250 per month per cycle.

BITS Pilani, K K Birla Goa Campus

The charge is made only on the number of cycle in storage on the last date of a month. The company does not have any beginning inventory and It does not wish to have any ending inventory. Formulate the LPP. Formulation: Decision Variables: x1 = number of cycles produced in October x2 = number of cycles produced in November x3 = number of cycles produced in December
BITS Pilani, K K Birla Goa Campus

In October: x1 5000, Cost of production = 2000 x1 Inventory at the end of October: S1=x1-5000 Cost of Storage = 250 S1 In November: Production is x2, therefore Cost of production = 2000 x2 We have S1+x210000 Inventory at the end of November: S2= S1+x2-10000 Cost of Storage = 250 S2

BITS Pilani, K K Birla Goa Campus

In December production x3: Cost of production = 2000 x3 We have S2+x3=15000 (no ending inventory) Objective Function: Minimize the total cost = Cost of production + Cost of storage Min z = 2000( x1+x2+x3) + 250(S1+S2) Subject to x1- S1 = 5000 x2+S1-S2 =10000 x3 +S2=15000 0 xi 5000, i =1,2,3 S10, S20
BITS Pilani, K K Birla Goa Campus

Example 6
A company has two grades of inspectors, I and II to undertake quality control inspection. At least, 1500 pieces must be inspected in an 8-hour day. Grade I inspector can check 20 pieces in an hour with an accuracy of 96% and, grade II inspector can check 14 pieces in an hour with an accuracy of 92%. The wages of grade I inspectors are $5 per hour, while the wages of grade II inspectors are $4 per hour. An error made by an inspector cost $3 to the company. If there are, in all, 10 grade I inspectors and 15 grade II inspectors in the company, find the optimal assignment of inspectors that minimizes the daily inspection cost.

BITS Pilani, K K Birla Goa Campus

Formulation
Let x1 and x2 be the number of Grade I and II inspectors assigned by the company. Objective : minimize the daily cost of inspection Wages paid to the inspectors and the cost of their inspection errors. The cost of grade I inspector/hour is $(5 + 3 0.04 20) = $ 7.40 per hour The cost of grade II inspector/hour is $(4 + 3 0.08 14) = $ 7.36 per hour
BITS Pilani, K K Birla Goa Campus

The objective function is Minimize Z = 8(7.40 x1 + 7.36 x2) = 59.20 x1 + 58.90 x2 subject to x1 10 x2 15 20 8 x1 + 14 8 x2 1500. x1 0, x2 0

BITS Pilani, K K Birla Goa Campus

Therefore, LPP in final form Min. Z = 59.20 x1 + 58.90 x2 Subject to x1 10 x2 15 160 x1 + 112 x2 1500 x1 0, x2 0

BITS Pilani, K K Birla Goa Campus

Thanks

BITS Pilani, K K Birla Goa Campus