Scribd
Upload
218 views37 pages

Chapter 4 Duality and Post Optimal Analysis

The document discusses linear programming duality theory and concepts such as constructing the dual problem from the primal, relationships between optimal solutions and objective values of the primal and dual problems, and post-optimal analysis techniques like sensitivity analysis and evaluating the effects of changes to constraints or the objective function on the optimal solution. Examples are provided to illustrate dual problem formulation and using the primal-dual relationships to determine dual optimal values or evaluate the impact of parameter changes on feasibility and optimality.

Uploaded by

Mir Md Mofachel Hossain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
218 views37 pages

Chapter 4 Duality and Post Optimal Analysis

The document discusses linear programming duality theory and concepts such as constructing the dual problem from the primal, relationships between optimal solutions and objective values of the primal and dual problems, and post-optimal analysis techniques like sensitivity analysis and evaluating the effects of changes to constraints or the objective function on the optimal solution. Examples are provided to illustrate dual problem formulation and using the primal-dual relationships to determine dual optimal values or evaluate the impact of parameter changes on feasibility and optimality.

Uploaded by

Mir Md Mofachel Hossain
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 37

Probability and Statistics

Course Code: IPE 2203

Chapter 4
Duality and Post-Optimal Analysis

Instructor: Md. Rasel Sarkar, Asst. Professor


Dept. of IPE, RUET, Bangladesh
Academic Session: 2019-20
LP Dual
 The original LP is referred to as primal
 Every linear programming problem has associated with
its another linear programming problem called the dual.
 The relationships between the dual problem and the
primal prove to be extremely useful in a variety of ways.
 If one problem is unbounded, the other one is infeasible
 If one problem has an optimal solution, so does the
other, and the optimal values are the same
 Having an optimal solution for one of them, we can
determine a dual optimal solution for the other
Unrestricted (unconstrained)
Variables
Sometimes decision variables are unrestricted in sign (positive,
negative or zero). Let,

x1 = no. of units produced by machine 1


x2 = no. of worker hired/ fired in a company
x2 is unrestricted in sign

In all such cases, the decision variables can be expressed as the


difference between two non-negative variables.
Put,
where,
Construction of Dual from the
Primal
1. For each primal constraint assign a dual variable.
2. For each primal variable construct a dual
constraint.
3. The coefficients of a primal variable define the
left-hand side of the dual constraint and its
objective function coefficient defines the right-
hand side of that constraint.
4. The right-hand side of the primal constraints
constructs the objective function coefficients of
the dual.
5. One problem seeks maximization and the other
seeks minimization and vice versa.
6. An equality (=) constraint in the primal
corresponds to a dual variable that must be
unrestricted in sign.
7. Conversely, when a primal variable is unrestricted
in sign, its dual constraint must be in equation
form.
8. Dual of the dual is primal
9. The sense of optimization, direction of
inequalities, and the signs of the variables in the
dual are governed by the rules in following table.

Primal  Dual
Dual  Primal
Example - 1

y1
y2
y3
y4
# Example - 2
Primal problem

Equation form (Standard form)


Dual form

y2 unrestricted
# Example - 3
Primal
y1
y2
y3

Equation form
minimize z = ..
Subject to
x1 + 2x2 – x3 = 3
2x1 – 4x2 + x4 = 5
x1 – x2 =3
# Dual
# Example – 4
Primal

Equation form
Dual form

Subject to

y2 0
y3 0
y1 unrestricted
Simplex tableau layout
Inverse matrix:
The constraint coefficients under the starting basic
variables form an identity matrix
With this arrangement, subsequent iterations of
the simplex tableau modify the elements of the
identity matrix to produce what is known as the
inverse matrix.
# Starting Tableau

# ith iteration
Primal–Dual relationships
• Relationship with optimal solution
• Relationship between optimal values of the
primal and dual variables
• Relationship between objective values of the
primal and dual
• Simplex tableau Computations
Relationship between optimal values of
the primal and dual variables

This section provides methods for determining the


dual values –
# Example

Determination of dual values:


Optimal primal tableau:

Dual values:
Relationship between objective values of
the primal and dual
• For any pair of feasible primal and dual solutions

• At the optimum, the relationship holds as a strict


equation, meaning that the two objective values
are equal.
# Example

(x1 = 0, x2 = 0, x3 = 8/3) and (y1 = 6, y2 = 0) (arbitrary) are feasible primal


and dual solutions.
The objective functions are –

Since z < w, the solutions are not optimal.


Simplex tableau Computations

1. Computations of Constraint columns


In any simplex iteration, a left-hand or a right-hand
side column is computed as follows:
# Example

Optimal tableau
we have

Similar computations generate the optimal columns


for x2, x3, x4, R, and the right-hand side (verify!).
2. Computations of Objective z-row
In any simplex iteration, the objective equation
coefficient (reduced cost) of xj is computed as
follows:
# Example
Optimal tableau

Objectives values:
The optimal values of the dual variables, (y1, y2) = (29/ 5, - 2/5)

Similar computations can be used to determine the z-


coefficients of x2, x3, and x4 (verify!).
Post-optimal analysis
Sensitivity analysis: determining the ranges for the
different LP parameters that keeps the optimum
solution unchanged.
Post-optimal analysis: deal with making changes in
the parameters of the model and finding the new
optimum solution.
Changing the
constraints
Post-optimal
analysis
Changing the
objective function
Changes affecting feasibility
(Changing the constraints)
Two cases:
- change in right-hand side of the constraints
- adding a new constraint to the model
In both cases, infeasibility occurs when one or
more of the current basic variables become
negative.
TOYCO model
TOYCO uses three operations to assemble three
types of toys—trains, trucks, and cars. The daily
available times for the three operations are 430, 460,
and 420 mins, respectively, and the revenues per unit
of toy train, truck, and car are $3, $2, and $5,
respectively. The assembly times per train at the
three operations are 1, 3, and 1 mins, respectively.
The corresponding times per truck and per car are (2,
0, 4) and (1, 2, 0) mins (a zero time indicates that the
operation is not used). The objective is to maximize
profit.
Example: Consider the TOYCO model -

Optimum tableau
Changes in the right-hand side

Situation 1.
Suppose that TOYCO is increasing the daily capacity
of operations 1, 2, and 3 to 600, 640, and 590
minutes,
Thus, the new basic solution is computed as follows:
The current basic variables x2, x3, and x6 remain feasible
at the new values 140, 320, and 30 units, respectively.
The associated optimum revenue is $1880.

Situation 2.
Another proposal shifts the slack capacity of operation 3
(x6 = 20 minutes) to the capacity of operation 1.
The resulting solution is –

The resulting solution is infeasible because x6 = -40,


Addition of a new constraint

Situation 1.
TOYCO is considering addition of a fourth assembly
operation.
The daily capacity of the new operation is 500
minutes and the times per unit for the three
products on this operation are 3, 1, and 1 minutes,
respectively.
The new constraint for operation 4 is
This constraint is redundant because it is satisfied by
the current optimum solution x1 = 0, x2 = 100, and x3
= 230.
Hence, the current optimum solution remains
unchanged.

Situation 2.
Suppose the fourth operation are changed to 3, 3,
and 1 minutes, respectively.
The new constraint for operation 4 is
This constraint is not satisfied by the current
optimum solution; namely, for x1 = 0, x2 = 100, and x3
= 230
x7 = 500 - 13 * 0 + 3 * 100 + 1 * 2302 = -30
This means that the new constraint is not redundant.
Changes affecting optimality
(Changing the objective function)
Changes in the objective function coefficients

You might also like