IE2110 Operations Research I

Dr. Chew Ek Peng Office: E1A-06-06 Tel: 6516-6554 Email: isecep@nus.edu.sg Dr. Ng Kien Ming Office: Tel: 6516-5541 Email: isenkm@nus.edu.sg

Topics

Introduction to Operations Research Linear Programming

Model Formulation and Graphical Solution Properties of LP and Simplex Approach Computer Solution using Excel Duality Sensitivity Analysis Modeling

Topics

Transportation Models Network Models NonLinear Programming (NLP)

Textbook

Introduction to Mathematical Programming (Operations Research: Volume One), Wayne L. Winston & Munirpallam Venkataramanan, Thomson, 4th edition IVLE
Lecture note Assignments

Assessment

Assignments: 10% Term Paper: 15% Midterm Examination: 15% (5th October) Final Examination: 60%

Operations Research
Scientific approach to managerial decision making that seeks to best design and operate a system, usually under conditions requiring the allocation of scarce resources.
A system is an organization of interdependent components that work together to accomplish the goal of the system
6

Origins of Operations Research

World War II

allocate scarce resources to various military operations and to activities within each operations complexity and specialization in organizations

Industrial after WW II

Battle of the Buldge

On December 16, 1944, in the last year of World War II, two German panzer armies, supported by a third army of infantry, together totaling more than 250,000 men, staged a massive counteroffensive in northern France, overwhelming the American First Army in the Ardennes. The offensive emanated from the German defensive line along the Our River, north of the city of Luxembourg, and was directed almost due west toward Namur and Liege in Belgium. The result, after several days of fighting, was a huge "bulge" in the Allied line and, therefore, this last major ground battle of World War II became known as the Battle of the Bulge.
8

On December 20, General Dwight D. Eisenhower, Supreme Allied Commander, called on General George Patton to attack the German offensive with his Third Army, which was then situated near Verdun, approximately 100 miles due south of the German left flank. Patton's immediate objective was to relieve the 101st Airborne and elements of Patton's own 9th and 10th Armored Divisions surrounded at Bastogne. Within 48 hours, on December 22, Patton was able to begin his counteroffensive, with three divisions totaling approximately 62,000 men. The winter weather was cold with snow and fog, and the roads were icy, making the movement of troops, tanks, supplies, and equipment a logistical nightmare. Nevertheless, on December 26 Bastogne was relieved, and on January 12, 1945, the Battle of the Bulge effectively ended in one of the great Allied victories of the war.
9

10

Phases of Operations Research

Defining the Problem Observe the System Formulate the Model Verify the Model Model Solution Select Suitable Alternatives Present the Results Implementing the Results

11

An Introduction to Modeling
A Modeling Example Eli Daisy produces the drug Wozac in huge batches by heating a chemical mixture in a pressurized container. Each time a batch is produced, a different amount of Wozac is produced. The amount produced is the process yield (measured in pounds). Daisy is interested in understanding the factors that influence the yield of Wozac production process. The solution on subsequent slides describes a model building process for this situation.

Daisy is interested in determining the factors that influence the process yield. This would be referred to as a descriptive model since it describes the behavior of the actual yield as a function of various factors. Daisy might determine the following factors influence yield: Container volume in liters (V) Container pressure in milliliters (P) Container pressure in degrees centigrade (T) Chemical composition of the processed mixture

Letting A, B, and C be the percentage of the mixture made up of chemical A, B, and C, then Daisy might find , for example, that:

Yield = 300 + 0.8V +0.01P + 0.06T + 0.001T*P - 0.01T2 0.001P2 + 11.7A + 9.4B + 16.4C + 19A*B + 11.4A*C 9.6B*C

To determine this relationship, the yield of the process would have to measured for many different combinations of the previously listed factors. Knowledge of this equation would enable Daisy to describe the yield of the production process once volume, pressure, temperature, and chemical composition were known.

Prescriptive or Optimization Models

Prescriptive models prescribes behavior for an organization that will enables it to best meet its goals. Components of this model include: objective function(s) decision variables constraints An optimization model seeks to find values of the decision variables that optimize (maximize or minimize) an objective function among the set of all values for the decision variables that satisfy the given constraints.

The Objective Function The Daisy example seeks maximize the yield for the production process. In most models, there will be a function we wish to maximize or minimize. This function is called the models objective function. To maximize the process yield we need to find the values of V, P, T, A, B, and C that make the yield equation (below) as large as possible.
Yield = 300 + 0.8V +0.01P + 0.06T + 0.001T*P - 0.01T 0.001P + 11.7A + 9.4B + 16.4C + 19A*B + 11.4A*C 9.6B*C
2 2

In many situations, an organization may have more than more objective. For example, in assigning students to the two high schools in Bloomington, Indiana, the Monroe County School Board stated that the assignment of students involve the following objectives: equalize the number of students at the two high schools minimize the average distance students travel to school have a diverse student body at both high schools

The Decision Variables

Variables whose values are under our control and influence system performance are called decision variables. In the Daisy example, V, P, T, A, B, and C are decision variables.

Constraints In most situations, only certain values of the decision variables are possible. For example, certain volume, pressure, and temperature conditions might be unsafe. Also, A, B, and C must be nonnegative numbers that sum to one. These restrictions on the decision variable values are called constraints.

Suppose the Daisy example has the following constraints: 1. Volume must be between 1 and 5 liters 2. Pressure must be between 200 and 400 milliliters 3. Temperature must be between 100 and 200 degrees centigrade 4. Mixture must be made up entirely of A, B, and C 5. For the drug to perform properly, only half the mixture at most can be product A.

Mathematically, these constraints can be expressed:

V5 V1 P 400 P 200 T 200 T 100 A0 B0 C0 A + B + C = 1.0 A 0.5

The Complete Daisy Optimization Model

Letting z represent the value of the objection function (the yield), the entire optimization model may be written as: maximize z = 300 + 0.8V +0.01P + 0.06T + 0.001T*P - 0.01T
2 2

0.001P + 11.7A + 9.4B + 16.4C + 19A*B + 11.4A*C 9.6B*C Subject to (s.t.) V5 V1 P 400 P 200 T 200 T 100 A + B + C = 1.0 A 0.5 A0 B0 C0

Any specification of the decision variables that satisfies all the models constraints is said to be in the feasible region. For example, V = 2, P = 300, T = 150, A = 0.4, B = 0.3 and C = 0.3 is in the feasible region. An optimal solution to an optimization model any point in the feasible region that optimizes (in this case maximizes) the objective function.

Problem Set 1(MultiPeriod Production Smoothing Model)

A company will manufacture a product for the next four months: March, April, may and June. The demands for each month are 520, 720, 520, and 620 units, respectively. The company has a steady workforce of 10 employees but can meet fluctuating production needs by hiring and firing temporary workers, if necessary. The extra costs of hiring and firing in any month are $200 and $400 per worker, respectively. A permanent worker can produce 12 units per month, and a temporary worker, lacking comparable experience, only produce 10 units per month. The company can produce more than needed in any month and carry the surplus over to the succeeding month at a holding cost of $50 per unit per month. Develop an optimal hiring/firing policy for the company over the four month planning horizon.
24

Problem Set 2 (Security Force Scheduling)

A personnel manager must schedule the security force in such a way as to satisfying requirements shown in the following table:
Time Midnight 4am 4am 8am 8am Noon Noon 4pm 4pm 8pm 8pm Midnight Minimum number of officers required 5 7 15 7 12 9

25

(Cont.)

Officers work 8 hour shifts. There are 6 such shifts each day. The starting and ending times for each shift are given as table below:
Shift 1 2 3 4 5 6 Starting Time Midnight 4am 8am Noon 4pm 8pm Ending Time 8am Noon 4pm 8pm Midnight 4am
26

Types of Models
Static and Dynamic Models
A static model is one in which the decision variables do not involve sequences of decisions over multiple periods. A dynamic model is a model in which the decision variables do involve sequences of decisions over multiple periods. In a static model, we solve a one shot problem whose solutions are prescribe optimal values of the decision variables at all points in time. The Daisy problem is an example of a static model.

Static and Dynamic Models

For a dynamic model, consider a company (SailCo) that must determine how to minimize the cost of meeting (ontime) the demand for sail boats it produces during the next year. SailCo must determine the number of sail boats to produce during each of the next four quarters. SailCos decisions must be made over multiple periods and thus posses a dynamic model.

Linear and Nonlinear models

Suppose that when ever decision variables appear in the objective function and in the constraints of an optimization model the decision variables are always multiplied by constants and then added together. Such a model is a linear model. The Daisy example is a nonlinear model. While the decision variables in the constraints are linear, the objective function is nonlinear since the objective function terms:
0.001T*P, - 0.01T , 0.001P , 19A*B, 11.4A*C, and 9.6B*C
2 2

are nonlinear. In general, nonlinear models are much harder to solve.

Integer and Noninteger Models

If one or more of the decision variables must be integer, then we say that an optimization model is an integer model. If all the decision variables are free to assume fractional values, then an optimization model is a noninteger model. The Daisy example is a noninteger example since volume, pressure, temperature, and percentage composition are all decision variables which may assume fractional values. If decision variables in a model represent the number of workers starting during each shift, then clearly we have a integer model. Integer models are much harder to solve then noninteger models.

Deterministic and Stochastic Models

Suppose that for any value of the decision variables the value of the objective function and whether or not the constraints are satisfied is known with certainty. We then have a deterministic model. If this is not the case, then we have a stochastic model. If we view the Daisy example as a deterministic model, then we are making the assumption that for given values of V, P, T, A, B, and C the process yield will always be the same. Since this is unlikely, the objective function can be viewed as the average yield of the process for given decision variable values.

Classification of Management Science Techniques

Modeling Techniques
32

Business Use of Operations/Management Science

Some application areas: - Project Planning - Capital Budgeting - Inventory Analysis - Production Planning - Scheduling - Waiting Time Analysis - Distribution Network Design
33

Chapter 3 Linear Programming: Model Formulation and Graphical Solution

34

Chapter Topics

Model Formulation A Maximization Model Example Graphical Solutions of Linear Programming Models A Minimization Model Example Irregular Types of Linear Programming Models Simplex Approach Characteristics of Linear Programming Problems

35

Linear Programming: An Overview

Objectives of business decisions frequently involve maximizing profit or minimizing costs. Linear programming is an analytical technique in which linear algebraic relationships represent a firms decisions, given a business objective, and resource constraints. Steps in application:

Identify problem as solvable by linear programming. Formulate a mathematical model of the unstructured problem. Solve the model. Implementation
36

Model Components

Decision variables - mathematical symbols representing levels of activity of a firm. Objective function - a linear mathematical relationship describing an objective of the firm, in terms of decision variables - this function is to be maximized or minimized. Constraints requirements or restrictions placed on the firm by the operating environment, stated in linear relationships of the decision variables. Parameters - numerical coefficients and constants used in the objective function and constraints.

37

Summary of Model Formulation Steps

Step 1 : Clearly define the decision variables Step 2 : Construct the objective function Step 3 : Formulate the constraints

38

LP Model Formulation A Maximization Example (1 of 3)

Product mix problem - Beaver Creek Pottery Company How many bowls and mugs should be produced to maximize profits given labor and materials constraints? Product resource requirements and unit profit:

39

40

LP Model Formulation A Maximization Example (2 of 3)

Resource Availability: Decision Variables: Objective Function: Resource Constraints: Non-Negativity Constraints: 40 hrs of labor per day 120 lbs of clay x1 = number of bowls to produce per day x2 = number of mugs to produce per day Maximize Z = $40x1 + $50x2 Where Z = profit per day 1x1 + 2x2 40 hours of labor 4x1 + 3x2 120 pounds of clay x1 0; x2 0

41

LP Model Formulation A Maximization Example (3 of 3)

Complete Linear Programming Model: Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x2 + 3x2 120 x1, x2 0

42

Feasible Solutions
A feasible solution does not violate any of the constraints: Example x1 = 5 bowls x2 = 10 mugs Z = $40x1 + $50x2 = $700 Labor constraint check: 1(5) + 2(10) = 25 < 40 hours, within constraint Clay constraint check: 4(5) + 3(10) = 70 < 120 pounds, within constraint

43

Infeasible Solutions
An infeasible solution violates at least one of the constraints: Example x1 = 10 bowls x2 = 20 mugs Z = $1400 Labor constraint check: 1(10) + 2(20) = 50 > 40 hours, violates constraint

44

Graphical Solution of LP Models

Graphical solution is limited to linear programming models containing only two decision variables (can be used with three variables but only with great difficulty). Graphical methods provide visualization of how a solution for a linear programming problem is obtained.

45

Coordinate Axes Graphical Solution of Maximization Model (1 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Coordinates for Graphical Analysis

46

Labor Constraint Graphical Solution of Maximization Model (2 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Graph of Labor Constraint

47

Labor Constraint Area Graphical Solution of Maximization Model (3 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Labor Constraint Area

48

Clay Constraint Area Graphical Solution of Maximization Model (4 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Clay Constraint Area

49

Both Constraints Graphical Solution of Maximization Model (5 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Graph of Both Model Constraints

50

Definitions

A feasible solution is a solution for which all constraints are satisfied A feasible region is the collection of all feasible solution An optimal solution is a feasible solution that has the most favorable value of the objective function

51

Feasible Solution Area Graphical Solution of Maximization Model (6 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Feasible Solution Area

52

Objective Function Solution = $800 Graphical Solution of Maximization Model (7 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Objection Function Line for Z = $800

53

Alternative Objective Function Solution Lines Graphical Solution of Maximization Model (8 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Alternative Objective Function Lines

54

Optimal Solution Graphical Solution of Maximization Model (9 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Identification of Optimal Solution

55

Optimal Solution Coordinates Graphical Solution of Maximization Model (10 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Optimal Solution Coordinates

56

Extreme (Corner) Point Solutions Graphical Solution of Maximization Model (11 of 12)

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Solutions at All Corner Points

57

Optimal Solution for New Objective Function Graphical Solution of Maximization Model (12 of 12)

Maximize Z = $70x1 + $20x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Optimal Solution with Z = 70x1 + 20x2

58

LP Model Formulation A Minimization Example (1 of 5)

Two brands of fertilizer available - Super-Gro, Crop-Quick. Field requires at least 16 pounds of nitrogen and 24 pounds of phosphate. Super-Gro costs $6 per bag, Crop-Quick $3 per bag. Problem: How much of each brand to purchase to minimize total cost of fertilizer given following data ?

59

60

LP Model Formulation A Minimization Example (2 of 5)

Decision Variables: x1 = bags of Super-Gro x2 = bags of Crop-Quick The Objective Function: Minimize Z = $6x1 + 3x2 Where: $6x1 = cost of bags of Super-Gro $3x2 = cost of bags of Crop-Quick Model Constraints: 2x1 + 4x2 16 lb (nitrogen constraint) 4x1 + 3x2 24 lb (phosphate constraint) x1, x2 0 (non-negativity constraint)
61

LP Model Formulation and Constraint Graph A Minimization Example (3 of 5)

Minimize Z = $6x1 + $3x2 subject to: 2x1 + 4x2 16 4x1 + 3x2 24 x1, x2 0

Graph of Both Model Constraints

62

Feasible Solution Area A Minimization Example (4 of 5)

Minimize Z = $6x1 + $3x2 subject to: 2x1 + 4x2 16 4x1 + 3x2 24 x1, x2 0

Feasible Solution Area

63

Optimal Solution Point A Minimization Example (5 of 5)

Minimize Z = $6x1 + $3x2 subject to: 2x1 + 4x2 16 4x1 + 3x2 24 x1, x2 0

Optimum Solution Point

64

Problem Statement Example Problem No. 1 (1 of 3)

Hot dog mixture in 1000-pound batches. Two ingredients, chicken ($3/lb) and beef ($5/lb). Recipe requirements: at least 500 pounds of chicken at least 200 pounds of beef Ratio of chicken to beef must be at least 2 to 1. Determine optimal mixture of ingredients that will minimize costs.

65

Solution Example Problem No. 1 (2 of 3)

Step 1: Identify decision variables. x1 = lb of chicken in mixture (1000 lb.) x2 = lb of beef in mixture (1000 lb.) Step 2: Formulate the objective function. Minimize Z = $3x1 + $5x2 where Z = cost per 1,000-lb batch $3x1 = cost of chicken $5x2 = cost of beef
66

Solution Example Problem No. 1 (3 of 3)

Step 3: Establish Model Constraints x1 + x2 = 1,000 lb x1 500 lb of chicken x2 200 lb of beef x1/x2 2/1 or x1 - 2x2 0 x1, x2 0 The Model: Minimize Z = $3x1 + 5x2 subject to: x1 + x2 = 1,000 lb x1 50 x2 200 x1 - 2x2 0 x1,x2 0
67

Example Problem No. 2 (1 of 3)

Solve the following model graphically: Maximize Z = 4x1 + 5x2 subject to: x1 + 2x2 10 6x1 + 6x2 36 x1 4 x1, x2 0 Step 1: Plot the constraints as equations
Figure 2.23 Constraint Equations
68

Example Problem No. 2 (2 of 3)

Maximize Z = 4x1 + 5x2 subject to: x1 + 2x2 10 6x1 + 6x2 36 x1 4 x1, x2 0 Step 2: Determine the feasible solution space

Figure 2.24 Feasible Solution Space and Extreme Points

69

Example Problem No. 2 (3 of 3)

Maximize Z = 4x1 + 5x2 subject to: x1 + 2x2 10 6x1 + 6x2 36 x1 4 x1, x2 0 Step 3 and 4: Determine the solution points and optimal solution
Figure 2.25 Optimal Solution Point
70

Irregular Types of Linear Programming Problems

For some linear programming models, the general rules do not apply. Special types of problems include those with:

Multiple optimal solutions Infeasible solutions Unbounded solutions

71

Multiple Optimal Solutions Beaver Creek Pottery Example

Objective function is parallel to to a constraint line. Maximize Z=$40x1 + 30x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0 Where: x1 = number of bowls x2 = number of mugs

Figure 2.20 Example with Multiple Optimal Solutions

72

An Infeasible Problem

Every possible solution violates at least one constraint: Maximize Z = 5x1 + 3x2 subject to: 4x1 + 2x2 8 x1 4 x2 6 x1, x2 0

Figure 2.21 Graph of an Infeasible Problem

73

An Unbounded Problem

Value of objective function increases indefinitely: Maximize Z = 4x1 + 2x2 subject to: x1 4 x2 2 x1, x2 0

Figure 2.22 Graph of an Unbounded Problem

74

Simplex Approach

More than two variables Convert all constraint in standard form Standard form requires that all constraints be in the form of equations (equalities). Introduction of slack variables

75

Slack Variables

Standard form requires that all constraints be in the form of equations (equalities). A slack variable is added to a constraint or is subtracted from a constraint to convert it to an equation (=) . When a slack variable is added to a constraint, it typically represents an unused resource. When a slack variable is subtracted from a constraint, it typically represents an excess above a constraint requirement level A slack variable contributes nothing to the objective function value.
76

Beaver Creek Pottery Company

Maximize Z = $40x1 + $50x2 subject to: 1x1 + 2x2 40 4x1 + 3x2 120 x1, x2 0

Standard form Max Z = 40x1 + 50x2 + 0s1 + 0s2 subject to:1x1 + 2x2 + s1 = 40 4x1 + 3x2 + s2 = 120 x1, x2, s1, s2 0

Solutions at All Corner Points

77

Linear Programming Model: Standard Form

Max Z = 40x1 + 50x2 + 0s1 + 0s2 subject to:1x1 + 2x2 + s1 = 40 4x1 + 3x2 + s2 = 120 x1, x2, s1, s2 0 Where: x1 = number of bowls x2 = number of mugs s1, s2 are slack variables

Solution Points A, B, and C with Slack

78

Fertilizer Example Slack Variables

Minimize Z = $6x1 + $3x2 subject to: 2x1 + 4x2 16 4x1 + 3x2 24 x1, x2 0 Standard form Min Z = $6x1 + $3x2+ 0s1 + 0s2 subject to: 2x1 + 4x2 - s1 =16 4x1 + 3x2 - s2 = 24 x1, x2 ,s1, s2 0
Optimum Solution Point
79

Linear Programming Model: Standard Form

Minimize Z = $6x1 + $3x2 + 0s1 + 0s2 subject to: 2x1 + 4x2 s1 = 16 4x1 + 3x2 s2 = 24 x1, x2, s1, s2 0

Graph of Fertilizer Example

80

How to Convert an LP to Standard Form

If an LP has both and constraints, apply the previous procedures to the individual constraints. Consider the example below. Nonstandard Form max z = 20x1 + 15x2 s.t. x1 100 x2 100 50x1 + 35x2 6000 20x1 + 15x2 2000 x1, x2 > 0 s.t. x1 x2 50x1 + 35x2 20x1 + 15x2 Standard Form max z = 20x1 + 15x2 + s1 + s2 + s3 = 100 = 100 = 6000 - s4 = 2000

x1, x2, s1, s2, s3, e4 > 0

Standard LP Form
Suppose an LP with m constraints and n variables has been converted into standard form. The form of such an LP is: max ( or min) z = c1x1 + c2x2 + +cnxn s.t. a11x1 + a12x2 + + a1nxn =b1 a21x1 + a22x2 + + a2nxn =b2 . . . . . .

am1x1 + am2x2 + + amnxn =bm xi 0 ( i = 1,2, , n)

Basic Solution
If we define:

The constraints may be written as a system of equations Ax = b. Consider a system Ax = b of m linear equations in n variables (where n m). A basic solution to Ax = b is obtained by setting n m variables equal to 0 and solving for the remaining m variables. This assumes that setting the n m variables equal to 0 yields a unique value for the remaining m variables, or equivalently, the columns for the remaining m variables are linearly independent.

Basic and NonBasic Variables

To find a basic solution to Ax = b, we choose a set of n m variables (the nonbasic variables, or NBV) and set each of these variables equal to 0. Then we solve for the values of the n (n m) = m variables (the basic variables, or BV) that satisfy Ax = b. Different choices of nonbasic variables will lead to different basic solutions. Consider the basic solutions to the system of 2 equations shown to the right: The number of nonbasic variables = 3 2 = 1. Setting, for example, NBV = {x3} (as shown to the right, then BV = {x1, x2}. We can obtain the values for these basic variables by setting x3 = 0. Solving we find x1 = 2, x2 = 1. Thus, x1 = 2, x2 = 1, and x3 = 0 is a basic solution. x1 + x2 =3

- x2 + x3 = -1

x1 + x2

=3

- x2 + 0 = -1

Basic and NonBasic Variables

If NBV = {x1} and BV = {x2, x3} are chosen instead, the basic solution becomes x1 = 0, x2 = 3, and x3 = 2. If NBV = {x2} and BV = {x1, x3} are chosen instead, the basic solution becomes x1 = 3, x2 = 0, and x3 = -1. Some sets of m variables do not yield a basic solution. Consider the linear system shown to the right: If NBV = {x3} and BV = {x1, x2} the corresponding basic solution would be: x1 + 2x2 + x3 = 1 2x1 + 4x2 + x3 = 3 x1 + 2x2 = 1 2x1 + 4x2 = 3

Since this system has no solution, there is no basic solution corresponding to BV = {x1, x2}.

Properties of LP
Any basic solution in which all variables are nonnegative is called a basic feasible solution ( or bfs). For the basic solutions on the previous slides, x1 = 2, x2 = 1, x3 = 0 and x1 = 0, x2 = 3, x3 = 2 are basic feasible solutions, but the basic solution x1 = 3, x2 = 0, x3 = -1 fails to be a bfs (because x3 < 0). The following two theorems explain why the concept of a basic feasible solution is of great importance in linear programming: Theorem 1 The feasible region for any linear programming problem is a convex set. Also, if an LP has an optimal solution, there must be an extreme point of the feasible region that is optimal. Theorem 2 For any LP, there is a unique extreme point of the LPs feasible region corresponding to each basic feasible solution. Also, there is at least one bfs corresponding to each extreme point in the feasible region.

What is a convex set?

A set is said to be convex if and only if

87

Examples
y x x y x y

(a)

(b)

(c)

Convexity of a set. (a) convex set; (b) convex set (polyhedron); (c) non-convex set
88

Show that LP feasible region is a convex set

Ax < b or Ax > b or Ax = b

89

Linear Programming Model: Standard Form

The relationship between extreme points and basic feasible solutions outlined in Theorem 2. The LP (with slack variables) was:

Max Z = 40x1 + 50x2 + 0s1 + 0s2 subject to:1x1 + 2x2 + s1 = 40 4x1 + 3x2 + s2 = 120 x1, x2, s1, s2 0
Both inequalities are satisfied in the shaded area. The extreme points are of the feasible region are 0, A, B, and C.

Solution Points A, B, and C with Slack

90

Adjacent Basic Feasible Solutions

For any LP with m constraints, two basic feasible solutions are said to be adjacent if their sets of basic variables have m 1 basic variables in common. For example, the bfs corresponding to point A is adjacent to the bfs corresponding to and point These points share (m 1 = 2 - 1 = 1) one basic variable, x2. Points 0 (BV = {s1,s2}) and B (BV = {x1,x2}) are not adjacent since they share no basic variables. Intuitively, two basic feasible solutions are adjacent if they both lie on the same edge of the boundary of the feasible region.

Simplex Algorithm
General description of the simplex algorithm solving an LP in a maximization problem:
Step 1 Find a bfs to the LP. We will call this bfs the initial bfs. In general, the most recent bfs will be called the current bfs, so at the beginning of the problem, the initial bfs is the current bfs. Step 2 Determine if the current bfs is an optimal solution to the LP. If it is not, find an adjacent bfs that has a larger zvalue. Step 3 Return to Step 2, using the new bfs as the current bfs.

Algebraic Form of LP
Consider Max z = cx subject to Ax = b x0 Max Z = 40x1 + 50x2 subject to:1x1 + 2x2 + s1 = 40 4x1 + 3x2 + s2 = 120 x1, x2, s1, s2 0

x0

The above equation can be decomposed into basic and non-basic variables of the following Max z = cBxB + cN xN subject to BxB + NxN = b xB 0, xN = 0 Let x1 and s1 be NBV

xB 0, xN = 0
94

We can expressed the BV as a function of NBV BxB + NxN = b BxB = b - NxN xB = B-1b - B-1NxN Substituting into the objective function, we have z = cBxB + cN xN z = cB(B-1b - B-1NxN) + cN xN z = cBB-1b + (cN - cBB-1N)xN Overall the LP model is z = cBB-1b + (cN - cBB-1N)xN s.t. xB = B-1b - B-1NxN xB 0, xN = 0

The LP model z = cBB-1b + (cN - cBB-1N)xN s.t. xB = B-1b - B-1NxN xB 0, xN = 0

Let x1 and s1 be NBV

xB 0, xN = 0
96

The LP model z = cBB-1b + (cN - cBB-1N)xN s.t. xB = B-1b - B-1NxN xB 0, xN = 0 Let x1 and s1 be NBV

xB 0, xN = 0
97

xB 0, xN = 0

xB 0, xN = 0

98

Max z = cBB-1b + {j in N}(cj -zj)xj subject to xB = B-1b - B-1NxN xB 0, xN = 0 where -(cj -zj ) is sometimes known as the reduced costs

xB 0, xN = 0

Linear Programming Model: Standard Form

Max Z = 40x1 + 50x2 + 0s1 + 0s2 subject to:1x1 + 2x2 + s1 = 40 4x1 + 3x2 + s2 = 120 x1, x2, s1, s2 0 Where: x1 = number of bowls x2 = number of mugs s1, s2 are slack variables

Solution Points A, B, and C with Slack

100

Linear Programming
What are the assumptions made when applying LP models in the real world problems?

101

Assumptions of Linear Programming Models

Proportionality - The rate of change (slope) of the objective function and constraint equations is constant. Additivity - Terms in the objective function and constraint equations must be additive. Divisibility -Decision variables can take on any fractional value and are therefore continuous as opposed to integer in nature. Certainty - Values of all the model parameters are assumed to be known with certainty (non-probabilistic).

102

Properties of LP

Corner-Point Feasible Solution

If there is exactly one optimal solution, then it must be a corner-point feasible solution. If there are multiple optimal solutions, then at least two must be adjacent corner-point feasible solution There are only a finite number of corner-point feasible solutions If a corner-point solution has no adjacent corner-point feasible solutions that are better, then there are no better corner-point feasible solutions anywhere, i.e., it is optimal.

103

End of Chapter

104