A Conic Algorithm for the Group Minimization Problem

Abstract

A new algorithm for the group minimization problem (GP) is proposed. The algorithm can be broadly described as follows. A suitable relaxation of(GP) is defined, in which any feasible point satisfies the group equation but may have negative components. The feasible points of the relaxation are then generated in order of ascending costs by a variant of a well-known algorithm of Glover, and checked for non-negativity. The first non-negative point is an optimal solution of (GP). Advantages and disadvantages of the algorithm are discussed; in particular, the implementation of the algorithm (which can be easily extended so as to solve integer linear programming problems) does not require group arithmetics.