Optimization formulation Optimization methods help us find solutions to problem where we seek to find the best of something. This lecture is about how.

Slides:



Advertisements
Similar presentations
Young Modulus Example The pairs (1,1), (2,2), (4,3) represent strain (millistrains) and stress (ksi) measurements. Estimate Young modulus using the three.
Advertisements

Ordinary Least-Squares
Design of Experiments Lecture I
Survey of gradient based constrained optimization algorithms Select algorithms based on their popularity. Additional details and additional algorithms.
Lecture 3 Linear Programming: Tutorial Simplex Method
Linear Programming. Introduction: Linear Programming deals with the optimization (max. or min.) of a function of variables, known as ‘objective function’,
Copyright © 2006 The McGraw-Hill Companies, Inc. Permission required for reproduction or display. 1 One-Dimensional Unconstrained Optimization Chapter.
Optimization of thermal processes2007/2008 Optimization of thermal processes Maciej Marek Czestochowa University of Technology Institute of Thermal Machinery.
Introduction to Management Science
19 Linear Programming CHAPTER
Introduction to Management Science
Optimization in Engineering Design 1 Lagrange Multipliers.
OPTIMAL CONTROL SYSTEMS
Martin Burger Institut für Numerische und Angewandte Mathematik European Institute for Molecular Imaging CeNoS Total Variation and related Methods Numerical.
Economics 214 Lecture 37 Constrained Optimization.
Lecture 35 Constrained Optimization
Lecture outline Support vector machines. Support Vector Machines Find a linear hyperplane (decision boundary) that will separate the data.
Introduction to Management Science
Lecture 10: Support Vector Machines
Introduction to Management Science
9.4 – Solving Absolute Value Equations and Inequalities 1.
Linear Programming Old name for linear optimization –Linear objective functions and constraints Optimum always at boundary of feasible domain First solution.
Chapter 10 Real Inner Products and Least-Square (cont.)
Linear programming. Linear programming… …is a quantitative management tool to obtain optimal solutions to problems that involve restrictions and limitations.
Optimization formulation Optimization methods help us find solutions to problems where we seek to find the best of something. This lecture is about how.
Stevenson and Ozgur First Edition Introduction to Management Science with Spreadsheets McGraw-Hill/Irwin Copyright © 2007 by The McGraw-Hill Companies,
Principles of Computer-Aided Design and Manufacturing Second Edition 2004 ISBN Author: Prof. Farid. Amirouche University of Illinois-Chicago.
9-1 Copyright © 2010 Pearson Education, Inc. Publishing as Prentice Hall Multicriteria Decision Making Chapter 9.
Multicriteria Decision Making
Linear programming Lecture (4) and lecture (5). Recall An optimization problem is a decision problem in which we are choosing among several decisions.
Optimization formulation Optimization methods help us find solutions to problems where we seek to find the best of something. This lecture is about how.
STDM - Linear Programming 1 By Isuru Manawadu B.Sc in Accounting Sp. (USJP), ACA, AFM
Fin500J: Mathematical Foundations in Finance
Nonlinear Programming
Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display. 1 Chapter 15.
Chapter 9 - Multicriteria Decision Making 1 Chapter 9 Multicriteria Decision Making Introduction to Management Science 8th Edition by Bernard W. Taylor.
Introduction to Linear Programming BSAD 141 Dave Novak.
Model Driven DSS Chapter 9. What is a Model? A mathematical representation that relates variables For solving a decision problem Convert the decision.
1 Announcements: The deadline for grad lecture notes was extended to Friday Nov. 9 at 3:30pm. Assignment #4 is posted: Due Thurs. Nov. 15. Nov is.
Mathematical Models & Optimization?
2005MEE Software Engineering Lecture 11 – Optimisation Techniques.
Hosted By The Math Institute
Risk Analysis & Modelling
Chapter 1 - Fundamentals Inequalities. Rules for Inequalities Inequalities.
Optimization unconstrained and constrained Calculus part II.
Overview of Optimization in Ag Economics Lecture 2.
ZEIT4700 – S1, 2015 Mathematical Modeling and Optimization School of Engineering and Information Technology.
Operations Research The OR Process. What is OR? It is a Process It assists Decision Makers It has a set of Tools It is applicable in many Situations.
NONNEGATIVE MATRIX FACTORIZATION WITH MATRIX EXPONENTIATION Siwei Lyu ICASSP 2010 Presenter : 張庭豪.
Systems of Linear Equations in Two Variables. 1. Determine whether the given ordered pair is a solution of the system.
Optimization Optimization is the methodology for obtaining the best alternative from all possible ones. Unconstrained optimization has only an objective.
ZEIT4700 – S1, 2015 Mathematical Modeling and Optimization School of Engineering and Information Technology.
Calculus-Based Optimization AGEC 317 Economic Analysis for Agribusiness and Management.
Optimization in Engineering Design 1 Introduction to Non-Linear Optimization.
Optimization formulation Optimization methods help us find solutions to problems where we seek to find the best of something. This lecture is about how.
Economics 2301 Lecture 37 Constrained Optimization.
Linear Programming Chapter 1 Introduction.
University of Colorado at Boulder Yicheng Wang, Phone: , Optimization Techniques for Civil and Environmental Engineering.
Linear programming Lecture (4) and lecture (5). Recall An optimization problem is a decision problem in which we are choosing among several decisions.
Optimal Control.
Linear Inequalities in One Variable
The Simplex Method: Standard Minimization Problems
CS5321 Numerical Optimization
Systems of Linear Equations in Two Variables
Introduction Basic formulations Applications
Using WinQSB to solve Linear Programming Models
ECEN 460 Power System Operation and Control
Objectives Identify solutions of linear equations in two variables.
Calculus-Based Optimization AGEC 317
Optimization formulation
Presentation transcript:

Optimization formulation Optimization methods help us find solutions to problem where we seek to find the best of something. This lecture is about how we formulate the problem mathematically. In this course we make the assumption that we have choices and that we can attach numerical values to the ‘goodness’ of each choice. This is not always the case. We may have problems where the only thing we can do is compare pairs of alternatives and tell which one is better, but not by how much. Can you think of an example?

Young Modulus Example

Direct unconstrained formulation Minimize maximum difference eps=[1 2 4]; sig=[1 2 3] e=linspace(0.5,1,101); sigmodel=e'*eps; sigr=ones(101,1)*sig; diff=abs(sigr-sigmodel); maxdiff=max(diff'); plot(e,maxdiff); xlabel('E'); ylabel('maximum difference'); Non-smooth function Difficult to optimize numerically.

Constrained formulation To avoid a non-smooth function, a standard trick is to add a design variable b that bounds the difference, as well as two constraints Note that the objective function is equal to one design variable, and the other variable E appears only in the constraints. Not only is the new objective and constraints smooth, but they are also linear.

Standard notation The standard form of an optimization problem used by most textbooks and software is Standard form of Young’s modulus fit

Multiple objectives It is common to have multiple objectives. For fitting Young’s modulus, we may have also wanted to minimize the average absolute difference. avgdiff=mean(diff'); plot(e,maxdiff,avgdiff,'r') xlabel('E') legend('maxdiff','avgdiff','Location','north') Interval of interest is (0.75,0.83)

Column design example Vanderplaats, Numerical optimization techniques for engineering design, 3 rd edition.

Design formulation.

In praise of normalization The column design formulation has normalized constraints Improves understanding: When equal to 0.1 we know it is 10% violation. Improve numerical conditioning, because all constraints are of the same order of magnitude. For same reason, it is good to normalize design variables when they are not all of the same order, like diameter and thickness.