Amal Machtalay

From mean-field games to agentbased models (and back) through Markov Chain aggregation

Amal during the defence of her PhD thesis.

Summary

How can we model the complex systems around us? This is the question at the heart of Amal Machtalay’s research, supervised by Prof. Retnani (Morocco) and Prof. Kressner (EPFL).

Her work focuses on mathematical models capable of reproducing the behaviour of systems comprising a large number of agents. Among their practical applications is road traffic, where every driver and every vehicle interacts with the others. Amal has developed a model capable of representing different driver behaviours, including those of autonomous vehicles, by combining the individual and global scales.

Beyond traffic, the project is developing numerical methods to solve large-scale mathematical problems, with a very practical aim: to make calculations faster and more efficient. In particular, it explores techniques for simplifying complex models whilst maintaining their accuracy, as well as the use of parallel computing and artificial intelligence.

This research opens up new possibilities for better simulating and controlling complex systems in fields as varied as mobility, population dynamics and multi-agent systems.

Download the complete file

Publications / Scientific news

Machtalay A, Habbal A, Ratnani A, Kissami I, Computational investigations of a multi-class traffic flow model : Mean-field and microscopic dynamics, Transportation Research Part B : Methodological, Volume 195, 2025, 103196, ISSN 0191-2615.