Shadi Sharif Azadeh

Title of her presentation: « From Modeling to Learning Systems: How AI Is Reinventing Operations Research for Dynamic Decision Making »

Abstract:

Operations Research has traditionally provided the mathematical foundations for designing, optimizing, and controlling complex systems. Yet many contemporary decision environments such as, transport networks, shared mobility platforms, autonomous vehicle systems, and real-time service operations become increasingly dynamic, uncertain, data-rich, and behaviorally responsive. In this talk, we first present a coherent view of how AI is changing the role of OR from a primarily model-driven discipline into a learning-enabled decision science. Then, we provide two research avenues in the context of transport networks: i) sequential integration of AI and OR for prediction, optimization and learning, and ii) AI methods embedded in optimization algorithms. As proof of concept, applications such as, demand and supply management in multimodal mobility under uncertain resource availability, train scheduling, and operational planning of Modularized Autonomous Vehicles (MAV) are shared. We believe that AI should not be viewed as a substitute for OR, but as a methodological amplifier. It predicts uncertain system states, learns from operational feedback, improves algorithmic efficiency, and enables adaptive optimization in systems where static models alone are no longer sufficient.

Bio:

Shadi Sharif Azadeh is an associate professor at Civil Engineering and Geosciences faculty and co-director of SUM (Sustainable Urban Multi-modal Mobility) lab at TU Delft in the department of Transport & Planning. She is the PI of the ERC Consolidator Grant TRANSFORM (2026-2031) that designs a smart “estimate-then-optimize” framework for a robust multi timescale asset management of multimodal transport systems in an uncertain environment.

Her areas of expertise include multi-level decisions under uncertainty, pricing and assortment optimization, choice-based optimization and large-scale combinatorial optimization. Her research has been recognized by INFORMS through Early Career Award and best paper awards. She is an Associate Editor at Transportation Science, Transportation Research Part C: Emerging Technologies, and Nature Series (npj) Sustainable Mobility and Transport. She also serves as editorial board editor for Transportation Research Part B: Methodological. She is steering committee member of hEART, board member of pricing and revenue management Euro working group. As of January 2026, she serves as the president-elect of INFORMS Transportation Science and Logistics Society (TSL).