A decision support system for the seller's return problem in the product line design

In this paper we present a decision support system (DSS) for the seller's return problem in the product line design. We consider that buyers choose the product that gives them maximum utility. The product line is designed so that the total marginal return from the sales of the products is maximized. The DSS we are presenting, performs "what if analysis" for a given product line profile and finds good solutions using a genetic algorithm approach. We present a genetic algorithm based heuristic for solving the product line design problem using the seller's marginal return criterion. The new approach is compared with a recently developed beam search method on randomly generated problems. Our method seems to be substantially better in terms of CPU time. Also, the solutions found by our method are better than those found by the beam search method in comparable times.