Binary integer programming formulations for scheduling in market-driven foundries

Abstract

This paper describes the development and solution of binary integer formulations for production scheduling problems in market-driven foundries. This industrial sector is comprised of small and mid-sized companies with little or no automation, working with diversified production, involving several different metal alloy specifications in small tailor-made product lots. The characteristics and constraints involved in a typical production environment at these industries challenge the formulation of mathematical programming models that can be computationally solved when considering real applications. However, despite the interest on the part of these industries in counting on effective methods for production scheduling, there are few studies available on the subject. The computational tests prove the robustness and feasibility of proposed models in situations analogous to those found in production scheduling at the analyzed industrial sector. (C) 2010 Elsevier Ltd. All rights reserved.