Scribd
Upload
143 views12 pages

DWA Optimizations

This document discusses optimization concepts and linear programming. It defines optimization as finding the best set of simultaneous decisions to maximize an objective function subject to constraints. Linear programming is described as a special type of optimization where the objective and constraints are linear functions of the decision variables. An example linear program is provided to maximize profits subject to material constraints. Sensitivity analysis is discussed as examining the impact of changes to input parameters like constraint right-hand sides or objective function coefficients. Shadow prices represent the change in optimal value from a one unit increase in a constraint right-hand side.

Uploaded by

sannyzhou
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
143 views12 pages

DWA Optimizations

This document discusses optimization concepts and linear programming. It defines optimization as finding the best set of simultaneous decisions to maximize an objective function subject to constraints. Linear programming is described as a special type of optimization where the objective and constraints are linear functions of the decision variables. An example linear program is provided to maximize profits subject to material constraints. Sensitivity analysis is discussed as examining the impact of changes to input parameters like constraint right-hand sides or objective function coefficients. Shadow prices represent the change in optimal value from a one unit increase in a constraint right-hand side.

Uploaded by

sannyzhou
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 12

Optimization

Optimization concepts Linear Program


Conceptual formulation Sensitivity analysis

What is Optimization?
Finding the best set of simultaneous decisions for a particular objective when the decision set is limited by certain constraints The three conceptual pillars of optimization
Decisions/decision variables Objective a function of the decision variables Constraints Every constraint has a Left Hand Side (LHS) and a Right Hand Side (RHS) LHS is a function of the decision variables RHS is a constant that denotes some real world limitation

Linear Program
A special type of optimization model in which the objective and constraints satisfy the mathematical property of linearity In mathematical terms, the objective and constraints must be linear continuous functions of the decision variables Linear functions of A, B and C:

13A + 20B + 25C 0.5A + (2/3)B + 42.37C 13A2 + 20AB + 25C log(A) + cos(B) max(A,0) IF(A< 5,0,10)

Non-Linear functions of A, B and C:


Linear and Non-linear functions

f(A) Linear function of A

f(A) Non - Linear function of A A

An Example of Linear Program


max 13 A + 20 B + 25C (Total Contribution)

subject to

Objective Coefficients Constraint Right hand sides (RHS)

material 1: 5A + 15B +15C 480 material 2: 4A + 4B +4C 160 material 3 : 35A + 20B + 25C 1190
Constraint Coefficients

(nonnegativity) A, B, C 0
All choices of decisions satisfying the constraints ----- feasible set can we get a better/worse result if we have a bigger feasible set? better or same

SENSITIVITY ANALYSIS
Also called post-optimality analysis

We need to examine the effect of changes in the values of input parameters

SHADOW PRICE
Shadow price for a resource is the marginal value of that resource It gives the change in the optimal value of the objective function for a unit change in the availability of that particular resource (the constraint RHS)

Shadow price is valid for changes within the allowable range

An Example of Linear Program


max 13 A + 20 B + 25C (Total Contribution)

subject to
5A + 15B +15C 480 4A + 4B +4C 160 35A + 20B + 25C 1190 (nonnegativity) A, B, C 0
Shadow Price of this constraint: If RHS increases by 1, the change in

optimal value of objective function

Changes in Objective Coefficients


Does the optimal solution change when the objective coefficients change? Sensitivity analysis for changes in objective coefficients helps us examine the robustness of the base case optimal solution.

An Example of Linear Program


max 13 A + 20 B + 25C (Total Contribution)

subject to

Will optimal solutions be changed if the coefficient 13 becomes 14?

5A + 15B +15C 480 4A + 4B +4C 160 35A + 20B + 25C 1190 (nonnegativity) A, B, C 0

Non-linear Programming
Non-linear programming
Can not use sensitivity report
o Need to re-run the solver to get new optimal decisions and objective value, if there are changes in objective coefficient or RHS of constraints

11

IP (Integer Programming)
IP--- when the decision variables can only take integer numbers (special case: binary numbers 1/0) IP can also be LP (linear programming) Even when IP is LP, the sensitivity report in excel is no longer valid
Only choices of integer decision variables limits the good features of LP, e.g., shadow prices. Re-run the solver if the parameters change

You might also like