Abstract: Contextual Stochastic Optimization (CSO) has emerged as a way model and solve real-world decision-making problems by integrating Machine Learning (ML) with optimization under uncertainty.  Many CSO models consider contextual (or side) information that are separate from the decisions. However, in many real-word applications, the decisions of the optimization model significantly affect the uncertain parameters, hence becoming contexts themselves.  For example, pricing decisions impact the uncertain demand of a product.

In this talk, we first introduce a methodology for modeling decision‑dependent uncertainty using a residuals-based stochastic optimization approach. We use various ML methods to learn how decisions shape uncertainty and then build a distribution (and thereby ambiguity set) around these predictions using empirical residuals.  We form a distributionally robust optimization model using the Wasserstein distance, and investigate its theoretical guarantees, including asymptotic optimality and finite sample results.  The resulting model is computationally challenging. Therefore, we devise a specialized Benders decomposition algorithm with nonlinear cutting planes to solve the resulting model.  We demonstrate the effectiveness of our proposed approach on a pricing and shipment planning problem.

Bio: Güzin Bayraksan is a Professor and Associate Chair for Research in the Integrated Systems Engineering Department at the Ohio State University, where she is also an affiliated faculty member of the Sustainability Institute and the Translational Data Analytics Institute. Her research interests are in optimization under uncertainty, particularly stochastic programming and distributionally robust optimization, using data-driven, contextual, and Monte Carlo simulation-based methods. She applies these models and methods to solve problems of critical societal interest in water and energy systems. Her papers have appeared in premier journals such as Mathematical Programming, SIAM Journal on Optimization and Operations Research, and her research has been supported by multiple grants from the National Science Foundation (NSF) and the Department of Energy (DoE). She is the recipient of NSF CAREER award, INFORMS ENRE Best Publication Award in Environment and Sustainability, Lumley Research Award (OSU), Five Star Faculty Award (UA), and the INFORMS Best Case Study award. She served as the Chair of Stochastic Programming Society (SPS) (2019–2023), the Vice Chair of Optimization under Uncertainty of INFORMS Optimization Society, and the President of the INFORMS Forum on Women in Operations Research and Management Science. She is currently serving or have served in the past on the editorial boards of Operations Research, INFORMS Journal on Optimization, IISE Transactions, Open Journal of Mathematical Optimization and Computational Optimization and Applications.