The Speakers

In this research we used mathematical optimization to inform strategic decisions surrounding the deployment of hydrogen in the Netherlands. One of the main challenges in this problem context is uncertainty about the future. The relevant time horizon extends from 2025 until 2050, and there is a lot of uncertainty regarding energy prices, governmental policies, technological development, etc. To address this uncertainty we developed and applied new methods for Robustness Analysis and Robust Optimization. 

We propose an online data compression approach for efficiently solving distributionally robust optimization (DRO) problems with streaming data while maintaining out-of-sample performance guarantees. Our method dynamically constructs ambiguity sets using online clustering, allowing the clustered configuration to evolve over time for an accurate representation of the underlying distribution. We establish theoretical conditions for clustering algorithms to ensure robustness, and show that the performance gap between our online solution and the nominal DRO solution can be written in terms of the distance between the true and compressed distributions. By varying the number of clusters, our method effectively balances robustness and online computational efficiency. We show that our analysis is compatible with well-established finite-sample and asymptotic guarantees for Wasserstein DRO, and provide additional dynamic regret bounds compared to online Wasserstein DRO with full information. Numerical experiments in mixed-integer portfolio optimization demonstrate significant computational savings, with minimal loss in solution quality. 

Justin Starreveld received his PhD from the University of Amsterdam, under the supervision of Prof. Dick den Hertog and Prof. Zofia Lukszo. His PhD research focuses on mathematical optimization under uncertainty, with an emphasis on applying such methods in practice. Prior to this, Justin obtained bachelor's and master's degrees in Econometrics from Erasmus University Rotterdam, where his passion for Operations Research was ignited. He currently works as an AI & Data Science Consultant at EyeOn, a Dutch consultancy firm that specializes in forecasting and supply chain planning.

Irina Wang is a PhD candidate in the department of Operations Research and Financial Engineering at Princeton University. Irina received a bachelor degrees in Operations Research and Information Engineering from Cornell University. Her research interests include robust optimization, decision-focused learning, optimization-based control, and stochastic multi-level optimization. She is the recipient of several honors and awards including a Princeton Wallace Memorial Fellowship, an INFORMS Computing Society Student Paper Award, and a Princeton School of Engineering and Applied Sciences Excellence Award.