Accession Number : AD0697320

Title :   TWO ALGORITHMS FOR INTEGER OPTIMIZATION,

Corporate Author : PURDUE UNIV LAFAYETTE IND KRANNERT GRADUATE SCHOOL OF INDUSTRIAL ADMINISTRATION

Personal Author(s) : Loehman,Edna ; Nghiem,Ph. T. ; Whinston,Andrew

Report Date : 05 SEP 1969

Pagination or Media Count : 32

Abstract : The paper presents two algorithms for solving zero-one programming problems. The algorithms are also suited for solving mixed integer programming problems, that is, when some of the variables are required to be zero or one but the other variables may take on any nonnegative value. The two algorithms are very similar in that both use the dual method to solve the associated linear programming problem. However, the tree search procedures used are different. In the first, the order of variables in a partial solution is fixed in backtracking. In the other, there is some flexibility in backtracking. Different choice procedures for selecting a variable to fix at zero or one are used. Computational results are presented. (Author)

Descriptors :   (*LINEAR PROGRAMMING, ALGORITHMS), OPTIMIZATION, SEARCH THEORY, COMPUTER PROGRAMMING

Subject Categories : Operations Research

Distribution Statement : APPROVED FOR PUBLIC RELEASE