Plenary Speakers

Prof. Dr. Dylan Possamaï

ETH Zürich

Title: Contract theory and energy markets

Abstract: We study how continuous-time contract theory can be used as a quantitative tool for the design of incentives in energy markets, focusing on two applications: demand–response management and pollution regulation. First, we model a demand–response program as a moral-hazard Principal–Agent problem in which a risk-averse consumer can exert costly effort to reduce both the mean and the volatility of her consumption deviations, while the producer observes consumption but not effort. We then extend the model to a large population with common shocks, leading to a mean-field Stackelberg setting where optimal incentives naturally combine (i) a “classical” component indexed on the individual consumption deviation and (ii) a benchmarking component indexed on aggregate behavior (equivalently, on the common noise). This yields explicit contract structures and comparative statics, and illustrates how controlling volatility can materially increase responsiveness. Second, we propose a regulator–producers model for pollution regulation in an electricity network. After an auction phase (cost functions given), the regulator chooses production and power flows subject to network constraints, and designs terminal transfers that induce producers to exert abatement effort affecting the drift of aggregate emissions. The set of implementable remunerations is characterized via a tractable dynamic representation, while the regulator’s problem leads to an HJB characterization and computable optimal contracts. Numerical experiments on a calibrated three-node network (inspired by the Chilean market) show that optimal contracts can substantially curb emissions, reducing pollution increments by more than 30% in the reported simulation study.

Prof. Dr. Kenan Zhang

EPFL

Title: Markov population potential games: An alternative framework for Markovian traffic assignment

Abstract: Traffic assignment models describe how selfish individuals interact on a physical network under prescribed behavioral rules, and how an equilibrium state emerges from these interactions. Although originally developed for long-term travel demand forecasting, their applications extend far beyond road traffic prediction. Among existing models, Markovian traffic assignment (MTA) strikes a balance between model representational power and analytical tractability by assuming memoryless (Markovian) system dynamics and state-dependent decision making. In this talk, I will introduce a new framework for Markovian traffic assignment, namely, Markov population potential game (MPPG), which integrates Markov game, potential game, and population game. By allowing introducing interdependent state transitions, MPPG generalizes classical MTA and substantially broadens the scope of applicability. A key advantage of MPPG, as with potential games in general, is the existence of an equivalent optimization problem whose optimal solution corresponds to the equilibrium. I will present sufficient conditions under which such a potential exist and can be constructed in closed form, along with solution algorithms based on the corresponding optimization formulation, then conclude with several applications of MPPG in transportation.

Manuel Bröchin

SBB

Dr. Philipp Germann

SBB

Florian Flückiger

SBB

Title: Application of OR at SBB

Abstract: From planning timetables at SBB Infrastructure, planning train driver shifts at SBB Passenger Division to optimizing freight transport chains at SBB Cargo, OR is used productively at SBB across all divisions and time horizons. We present these three examples in their business context and show how they are currently being modelled and solved.

Dr. Oliver Strub

Mobiliar

Dr. Mario Gnägi

BKW

Title: Applications of OR at Mobiliar and BKW

Abstract:

Mobiliar: We investigate the problem of optimizing the Strategic Asset Allocation (SAA) of the Mobiliar Insurance Group with respect to return and risk as measured by the Swiss Solvency Test (SST). The SST is a regulatory framework used by the Swiss Financial Market Supervisory Authority (FINMA) to assess the solvency of Swiss insurance companies. The key challenge is that evaluating the risk of a given SAA admits no closed-form solution, but instead requires running a simulation using a FINMA-provided tool with a runtime of approximately five minutes per evaluation. To address this, we apply a Genetic Algorithm in which FINMA's simulation tool serves as the fitness function. Using this approach, the expected return can be increased by more than 25 basis points without reducing the SST ratio, or alternatively the SST ratio can be improved by approximately 60 percentage points at unchanged expected returns.

BKW: The energy transition poses substantial challenges for BKW, in particular for planning and operating the distribution grid. Decentralized photovoltaic generation and the electrification of heating and mobility drive both bidirectional power flows and higher peak loads, pushing the distribution grid to its limits. As a result, distribution grid planning must become more data-driven and proactive to ensure security of supply while keeping investments economically reasonable. To tackle these challenges, our team at BKW develops and applies intelligent decision-support approaches for distribution grid planning, and we see strong potential in Operations Research to support this work. One concrete application is estimating available capacity in underground conduit systems. Feasible capacity is constrained by existing occupancy along the full route, engineering rules (e.g., reserve-capacity policies), and imperfect asset data. We outline how OR-based modeling can translate heterogeneous asset information into actionable capacity estimates and planning recommendations. Beyond conduit capacity assessment, we see further opportunities to apply Operations Research in distribution grid planning. We close with an outlook on related planning problems that we would like to tackle next by leveraging the full potential of Operations Research.