Multi-Stage\nStochastic Programming Under Endogenous\nUncertainty of Integrated Sustainable Chemical Process Design and\nExpansion Planning

Effective\ndecision-making is essential for minimizing\nthe environmental\nfootprint while strengthening the competitiveness of the chemical\nindustry. This paper proposes a multistage stochastic programming\n(MSSP) framework to take into account the design and expansion planning\nsimultaneously for a sustainable multinetwork that integrates the\nwater treatment, renewable energy supply, and carbon capture, utilization\nand storage (CCUS) with a main chemical processing sector under endogenous\nuncertainty related to conversion rates arising from the possible\nprocessing steps. Both economic and environmental concerns are involved\nin the optimization model for the entire system. The resulting mixed-integer\nlinear programming (MILP) model is then solved using the Lagrangean\ndecomposition algorithm. The proposed framework is further implemented\nin xylitol process design problems. Four cases are established based\non different combinations of uncertain parameter distributions and\ntime horizons, with model complexity increasing sequentially. The\neffectiveness of the proposed framework is further validated under\nnew randomized sampling scenarios. The results indicate that integrating\nmultinetworks can significantly reduce carbon emissions, thereby mitigating\nenvironmental impacts while satisfying production demands. Specifically,\ncarbon dioxide (CO<sub>2</sub>) can be fully captured and optimally\nutilized, wastewater can be completely treated, and economic benefits\ncan be effectively maximized. In comparison to the deterministic model,\nthe MSSP counterpart offers a sustainable, robust, and reliable solution\nfor integrated design and expansion planning over the specified time\nhorizon. Additionally, it achieves the value of stochastic solution\n(VSS) of 2%, potentially saving millions of dollars in long-term capacity\nplanning, thereby underscoring the advantages of incorporating endogenous\nuncertainty into the problem formulation.