Design and Scheduling of Chemical Batch Processes: Generalizing a Deterministic to a Stochastic Model

Abstract

A stochastic optimization model for the design and scheduling of batch chemical processes is developed in a Two-Stage
Stochastic Programming framework, with the uncertainty formulated through a number of discrete scenarios. The sparse model
presents binary variables in the first stage and systematically generalizes a deterministic model chosen from the literature, in an
approach based on computational complexity. The combination of single product campaign (SPC) with multiple machines was
found to be the most promising from a computational standpoint, and it is here generalized toward a stochastic environment within
the relaxation of the soft demand constraints. Numerical examples are presented, and the results point to a significant reduction of
8-20 % of the investment costs in comparison to the SPC non-relaxed case, without real losses if the multiple product campaign
(MPC) policy is adopted.