Accession Number : AD0770127

Title :   Concave Nonlinear Optimization Under Constraints with Integer Valued Variables.

Descriptive Note : Technical rept.,


Personal Author(s) : Loomis,Harlan W.

Report Date : NOV 1973

Pagination or Media Count : 49

Abstract : The branch and bound technique has been applied to solve both the mixed integer linear program, in which some but not necessarily all of the variables are restricted to be integers, and the concave nonlinear program having separable concave objective function and linear constraints. The subproblems for each of these branch and bound algorithms are linear programs with simple upper bounds. In the report, the method of penalties is applied to develop a new branching rule for the concave nonlinear program. This leads to a new branch and bound algorithm for the composite mixed integer, concave nonlinear program. The efficient solutions of the linear programming subproblems generated by these branch and bound algorithms is also discussed. (Author)

Descriptors :   (*Nonlinear programming, Optimization), Linear programming, Systems analysis, Algorithms

Subject Categories : Operations Research

Distribution Statement : APPROVED FOR PUBLIC RELEASE