Integration of planning, scheduling and control problems using data-driven feasibility analysis and surrogate models

In this work, a framework for the integration of planning, scheduling and control using data-driven methodologies is proposed. The framework consists of addressing the integrated problem as a grey-box optimization problem, and using data-driven feasibility analysis and surrogate models to approximate the unknown black-box constraints. We follow a systematic procedure to achieve this integration, consisting of two building blocks: first, we address the integration of scheduling and control followed by the integration of planning and scheduling. To handle dimensionality issues, we introduce the concept of feature selection when building the surrogate models. The methodology is applied to the optimization of an enterprise of air separation plants. (C) 2020 Elsevier Ltd. All rights reserved.