Single machine campaign planning under sequence dependent family setups and co-production

Abstract

We investigate tactical level production planning problem in process industries, with float glass manufacturing being the specific application domain. In the presence of high sequence dependent family setup costs, the need for planning production in batches, or campaigns as named in the float glass industry, arises. Moreover, the float glass manufacturing has some unique properties such as partially controllable co-production and uninterruptible production. The motivation of our work is a real life problem encountered at a major float glass manufacturing company in Turkey. We develop two mixed integer programming formulations and investigate some variants to solve the problem. Our formulations are capable of handling different resolutions in input data such as demand forecast expressed in discrete time and setup durations expressed in continuous time. We compare formulations both theoretically and by running computational experiments. Furthermore, we conduct additional experiments to gain insights about characteristics of the generated campaign plans from a business perspective.