Data-Driven Process Network Planning: A Distributionally Robust Optimization Approach

Abstract

Process network planning is an important and challenging task in process systems engineering. Due to the penetration of uncertainties such as random demands and market prices, stochastic programming and robust optimization have been extensively used in process network planning for better protection against uncertainties. However, both methods fall short of addressing the ambiguity of probability distributions, which is quite common in practice. In this work, we apply distributionally robust optimization to handling the inexactness of probability distributions of uncertain demands in process network planning problems. By extracting useful information from historical data, ambiguity sets can be readily constructed, which seamlessly integrate statistical information into the optimization model. To account for the sequential decision-making structure in process network planning, we further develop multi-stage distributionally robust optimization models and adopt affine decision rules to address the computational issue. Finally, the optimization problem can be recast as a mixed-integer linear program. Applications in industrial-scale process network planning demonstrate that, the proposed distributionally robust optimization approach can better hedge against distributional ambiguity and yield rational long-term decisions by effectively utilizing demand data information.