An approach for solving the lot-sizing problem of a market-driven foundry

Abstract

Foundries can be found all over Brazil and they are very important to its economy. In 2008, a mixed integer-programming model for small market-driven foundries was published, attempting to minimize delivery delays. We undertook a study of that model. Here, we present a new approach based on the decomposition of the problem into two sub-problems: production planning of alloys and production planning of items. Both sub-problems are solved using a Lagrangian heuristic based on transferences. An important aspect of the proposed heuristic is its ability to take into account a secondary practice objective solution: the furnace waste. Computational tests show that the approach proposed here is able to generate good quality solutions that outperform prior results.

Keywords:

• foundry
• lot-sizing
• heuristic

Acknowledgements

The authors thank the anonymous referees for their useful comments and suggestions. This research was partially funded by the Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (CNPq) and Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP).