A Local Search Approach for the Pattern Restricted One Dimensional Cutting Stock Problem

Abstract

As the setup cost of cutting patterns becomes more important in modern cutting industry, we consider the pattern restricted one dimensional cutting stock problem (1D-PRP), in which the number of stock rolls is minimized while the number of different cutting patterns is constrained within a bound given by a program parameter. For this problem, we propose a new heuristic algorithm based on local search, and incorporate a heuristic algorithm that provides a small subset of neighborhood which tends to contain good solutions in the original neighborhood. According to our computational experiments, the proposed algorithm attains a wide variety of good solutions which are comparable to the existing heuristic approaches for the one dimensional cutting stock problem (without pattern restriction).