Research Seminar - Francesca Maggioni

The Institute for Statistics and Mathematics is pleased to invite you to the next research seminar, which will take place on campus:

Francesca Maggioni ( Department of Management, Information, and Production Engineering (DIGIP), University of Bergamo)
Bounds and Approximations for Large-Scale Optimization Problems Under Uncertainty

Wednesday, November 4, 2026, at 5:15 p.m., in Building D4, Room D4.0.127.

Abstract: Many real-world decision problems are dynamic and subject to uncertainty. Stochastic programming offers a powerful approach for managing this uncertainty within a multi-period decision-making framework. However, as the number of stages increases, the computational complexity of these problems grows exponentially, posing significant challenges. To address this, approximation techniques are often used to simplify the original problem, providing useful upper and lower bounds for the optimal value of the objective function.
This talk explores methods for generating bounds for a wide variety of problem structures affected by uncertainty. We begin by discussing bounds based on scenario grouping under the assumption that a sufficiently large scenario tree is given but is unsolvable, both in the context of stochastic programming and distributionally robust optimization. Next, we extend these techniques to address more complex problems, including multi-horizon stochastic optimization and decision-dependent stochastic optimization.
Finally, the talk introduces the integration of these bounding methods with Benders’ decomposition. To reduce the computational burden of generating a cut for each scenario, a Benders refinement-chain cuts method is proposed, where scenario subsets are used to generate group-wise optimality cuts. This aggregation significantly reduces the number of cuts required, while preserving valid lower bounds. Theoretical relationships between cuts generated at different refinement levels are established.
Numerical experiments on various energy and transportation optimization problems demonstrate the efficiency of the proposed approaches.

For further information and the seminar schedule, please see:
https://www.wu.ac.at/en/statmath/research/research-seminar