Accession Number : AD0726027

Title :   Large Scale Convex Programming Models,

Corporate Author : RESEARCH ANALYSIS CORP MCLEAN VA

Personal Author(s) : McCormick,Garth P.

Report Date : JUL 1971

Pagination or Media Count : 9

Abstract : Almost all practical resource allocation programming problems which are convex can be represented as: min Summation over x of ((alpha sub ok)((1 sub ok)x)) subject to Summation over k of ((alpha sub ik)(L sub ik)x)) = or > 0, i = 1,...,m, where the alpha sub ik(.)) are convex functions of a single variable. In this paper the computational implications of such a representation are described. Some of the topics considered are: real world models in resource allocation which yield such structure, facility in preparing the input, simple methods of obtaining the first and second derivatives, use of the highly developed techniques for handling large scale linear programming problems on these structured convex programming problems, and second order algorithms for solving these problems. (Author)

Descriptors :   (*NONLINEAR PROGRAMMING, ALGORITHMS), LINEAR PROGRAMMING, PROBLEM SOLVING, COMPUTER PROGRAMMING, NUMERICAL ANALYSIS, OPTIMIZATION

Subject Categories : Operations Research

Distribution Statement : APPROVED FOR PUBLIC RELEASE