Data-driven scenario generation for two-stage stochastic programming

Optimisation under uncertainty has always been a focal point within the Process Systems Engineering (PSE) research agenda. In particular, efficient manipulation of large amount data for uncertain parameters constitutes crucial condition effectively tackling stochastic programming problems. this context, work proposes new data-driven Mixed-Integer Linear Programming (MILP) model Distribution & Moment Matching Problem (DMP). For cases with multiple copula-based simulation initial scenarios is employed as preliminary step. Moreover, integration clustering methods and DMP in proposed shown to enhance computational performance. Finally, we compare approach state-of-the-art scenario generation methodologies. Through number case studies highlight benefits regarding quality generated trees by evaluating corresponding obtained solutions. • A novel method hybridising mathematical sampling. distribution matching problem are modelled Mixed Integer optimisation problem. Case PSE literature indicate over approaches.