A research study submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology

Research Studies Project Report (M. Eng.) - Asian Institute of Technology, 1984

The search for a minimum sum of scrap production is the core requirement in any aluminum section manufacturing. this study develops a method for scrap reduction planning in an aluminum extruding factory which produces aluminum sections from aluminum billet. There are 81 different cross sections of aluminum sections under consideration. The main problems are what different billet sizes are used in producing those 81 cross sections of aluminum sections and what subset of billet sizes is selected to stock to supply total demand with the result of total minimum sum of scrap production in the factory. In developing the method, three successive stages of solution procedure are made, namely: (1) determine all feasible billet sizes that can be used to produce each of 81 cross sections of aluminum sections, (2) determine an optimum set of billet sizes used to produce those 81 cross sections; i.e. the set that gives minimum sum of scrap in extruding process, and (3) select a subset of billet sizes from the optimum set to stock; i.e. the set that gives a minimum sum of increased scrap production apart from the sum in extruding process. A computer program is employed for stage 1, zero-one integer linear programming is used for stage 2, and finally a dynamic programming is developed for stage 3, Results of the developed method indicate the list of optimum 81 billet sizes for those 81 cross sections of aluminum sections and the list of billet sizes, with different numbers of stocked sizes, selected to stock. They also indicate the number of each size needed to stock and the total sum of scrap produced in the factory. The study provides the analysis when the forecasted demand changes and the number of stocked sizes is varied.