Risk conscious scheduling of batch processes

We consider real-time scheduling problems of flexible batch processes under the special consideration of uncertainties. Any decision has to be made subject to a certain risk since it affects the future evolution of the process, which is not precisely predictable. This difficulty can be faced by a moving horizon approach with frequent re-optimisations. We propose the use of a model framework from stochastic programming to reflect the uncertainty and the potential of recourses realistically. The framework is applied to a real-world process from the polymer industries, a decomposition algorithm is sketched and numerical results are given.