A majority of all real world optimization problems are on too large of a scale to compute analytically so numerical methods are necessary to computationally solve them. MATLAB has an optimization toolbox with many functions that are able to handle combinations of linear, nonlinear, constrained, and unconstrained problems. These functions all have underlying algorithms used to help them to work so in this paper we will be going through the linear and nonlinear interior point method, golden search method, parabolic interpolation, newton's method, and quasi-newton methods within the functions linprog, fminbnd, fminunc, and fmincon. Also there are examples showing the implementation of these algorithms along with their corresponding analytic solutions if they are available. Finally, there is a research question in progress that we assess, and discuss an approach that can be used to solve.

Creator

• Robinson, Lynnette R.

Contributors

• Tilley, Burt S.

Degree

• MS

Unit

• Mathematical Sciences

Publisher

• Worcester Polytechnic Institute

Identifier

• etd-3931

Advisor

• Tilley, Burt S.

Defense date

• 2020-05-18

Year

• 2020

Date created

• 2020-05-18

Resource type

• Report

Rights statement

• In Copyright