Hardboard factories transform eucalyptus into rectangular plates by means of processes of disintegration, pressing and drying. These plates are then cut into smaller rectangular plates to satisfy customer demands. The scheduling of the cutting process is an important activity of the production planning and control of these companies due to the high costs related to trim losses. In this paper we present approaches to generate cutting patterns that minimize the waste of material, satisfying the constraints of the cutting equipment and the customer demands. We propose an algorithm based on dynamic programming, which can be combined with simple greedy constructive heuristics or the simplex primal algorithm with column generation. A case study was carried out in a large hardboard company located in Sao Paulo State, whose cutting process involves a technology with high degree of automation. The results show that the approaches have potential to produce good solutions compared to the ones utilized by the company.
cutting problems; dynamic programming; hardboard industry
