Lot sizing and scheduling in a large automated foundry

This work consists of a case study in a large foundry, which has three molding lines and two sets of furnaces. The molding lines produce different types of items, which must be made of different alloys produced by the furnaces. In each period, the production programming in this foundry has two important levels of decision making which are linked: 1) which alloys should be produced in the furnace in each period and 2) the quantity of each item to be produced in each molding machine. Initially, the productive process is described and then a mixed integer programming optimization model is proposed in which the lot-sizing of the items and the capacity programming of the furnaces are considered. A commercial package/software was utilized to obtain the computational results. A simulation lasting 5 working days was carried out and the computational results obtained from the model solution were compared with the ones obtained by the company, showing a significant increase in the productivity and reduction concerning delays of the delivery deadline.