Risk Management in the Scheduling of Batch Plants under Uncertain Market Demand

Abstract

A stochastic optimization approach is presented to manage risk in the short-term scheduling of multiproduct batch plants with uncertain demand. The problem is modeled using a two-stage stochastic optimization approach accounting for the maximization of the expected profit. The model is also extended to incorporate the availability of option contracts, thus providing significant flexibility within the uncertain environment. Management of risk is explicitly addressed by adding a control measure as a new objective to be considered, thus leading to multiobjective optimization formulations. Three alternative methodologies are assessed and compared. The importance of considering the uncertainty not only in the decision-making process but also in the control of the variability of outcomes is highlighted. Parametric solutions appealing to decision makers with different attitudes toward risk are obtained.