Integration between Dynamic Optimization and Scheduling of Batch Processes under Uncertainty: A Back-Off Approach

Abstract

The aim of this study is to present a decomposition algorithm that employs a new back-off methodology to consider stochastic-based parameter uncertainty for the integration of dynamic optimization and scheduling of multi-unit batch plants. This is achieved by solving a series of optimization problems involving scheduling and control decisions. Simulations based on Monte Carlo sampling techniques are used to propagate uncertainty into the system and determine back-off terms for the process operational constraints. At each step in the algorithm, back-off terms are updated such that the system moves away from the nominal solution until a convergence criterion is met, obtaining a solution that satisfies constraints up to a user-defined probability limit. This algorithm, when applied to a multi-product multi-unit batch plant, produces an optimal schedule and control profiles that remain dynamically feasible in the presence of stochastic-based uncertain parameters.