Global Algorithm for Generalized Affine Multiplicative Programming Problem

In this paper, a new outer space rectangle branch and bound algorithm is proposed for globally solving generalized affine multiplicative programming problem. By applying the equivalent transformations and affine approximations of bilinear function, the initial generalized affine multiplicative programming problem can be reduced to a linear relaxed programming problem. By subsequently refine the initial outer space rectangle, and by means of the subsequent solutions of a series of linear relaxed programming problems, the algorithm is convergent to the global optimal solution of the primal problem, and numerical results show the feasibility and effectiveness of the algorithm.