Integrated scheduling and dynamic optimization of grade transitions for a continuous polymerization reactor

Abstract

This paper presents a modeling and numerical solution method for an integrated grade transition and production scheduling problem for a continuous polymerization reactor. The optimal sequence of production stages and the transitions between them is supposed to be determined for producing a given number of polymer grades at certain amounts and quality specifications in the most economical way. The production schedule has to satisfy due dates for specific orders. This operational problem is cast into a mixed-integer dynamic optimization problem. Disjunctions and logical constraints are combined with a validated differential–algebraic model describing the polymer process during the production of a specific grade as well as along a transition between two different grades. The modeling and solution approach proposed by Oldenburg et al. [Oldenburg, J., Marquardt, W., Heinz D., & Leineweber, D. B. (2003)] is tailored to this problem class to provide an efficient solution technique. An industrial example process serves as an example to illustrate the modeling and solution techniques suggested.