Research Reports from the Department of Operations
Document Type
Report
Publication Date
4-1-1971
Abstract
Many nonlinear programming problems containing only upper and lower bounds are currently solved as unconstrained problems using transformation, or penalty methods of optimization. In this thesis it is shown that (1) Goldfarb's conjugate gradient algorithm can be simplified for NLP problems containing only upper and lower bounds, (2) the simplified algorithm requires less computer storage, and fewer multiplications per iteration with no loss in accuracy, and (3) the simplified algorithm is superior to transformation, or penalty methods when tested on published nonlinear test problems. A FORTRAN code for the nonlinear algorithm for upper and lower bounds, including complete documentation, is presented to provide a useful OR tool which could increase the number of successful applications of NLP techniques.
Keywords
Operations research, Nonlinear programming, Mathematical optimization, Computer storage devices, FORTRAN (Computer program language)
Publication Title
Technical Memorandums from the Department of Operations, School of Management, Case Western Reserve University
Issue
Technical memorandum no. 194
Rights
This work is in the public domain and may be freely downloaded for personal or academic use
Recommended Citation
Huckfeldt, Vaughn E., "Nonlinear Optimization with Upper and Lower Bounds" (1971). Research Reports from the Department of Operations. 339.
https://commons.case.edu/wsom-ops-reports/339