An Algorithm for Solving Lot Sizing and Cutting Stock Problem within Aluminum Fabrication Industry

Needless to say that the combined lot sizing and cutting stock problem is very relevant in a wide range of manufacturing industries. This paper presents an algorithmic solution approach to overcome the difficulty in solving non-linear integer formulation of the problem. The algorithm is based on the traditional approach where the lot sizing is determined for each period, and then the best cutting patterns are generated. However the algorithm adopts a backward lot-sizing based on available capacity and purchase cost burden index. Using 85 instances, the performance of the algorithm is assessed in terms of trim loss and computation time for a range of different piece lengths; number of final products; and number of scheduling periods. Good results are produced when compared with the practiced performance.