Workshop : Methods and tools for risk management of complex socio-technical systems

Main contact person : Jérémie Bosom, jeremie.bosom(at)ephe.sorbonne.fr

Persons in charge :
Marc Bui : marc.bui(at)ephe.sorbonne.fr
EPHE 4-14 rue Ferrus 75014 Paris
Marc Bui is full professor of Computer Science at the University of Paris 8 (since 1999) and at the Ecole
Pratique des Hautes Etudes (since 2005). His research interests include distributed computing, complex
systems modelling, distributed simulation with multi-agents systems and computational humanities.

Jérémie Bosom : jeremie.bosom(at)ephe.sorbonne.fr
EPHE 4-14 rue Ferrus 75014 Paris
Engineer, PhD candidate. in Computer Science at PSL University, Ecole Pratique des Hautes Etudes au
laboratoire CHArt. His research interests concern complex energy systems, AI and machine learning.

Summary

The aim of this session is to discuss methods and tools to contribute to risk management in complex socio-technical energy systems. One of the major challenges of the 21st century is to cope with the depletion of energy resources in face of an ever-increasing demand. Energy management and its associated risks fall within the domain of complex systems, indeed, from production to consumption, many subsystems are interdependent and exhibit possible failures such as power outage. Beyond physical and technical factors, human factors will also be now at the heart of such problems if small area autonomous eco-districts are promoted. The diculty of risk management of theses complex socio-technical systems stems from the complexity of modelling and are linked to the prediction of energy usages.

Agent-based modelling and simulation is a recent modelling paradigm. Many developments have been achieved by using it in various elds including social sciences, medical sciences, humanities or economics and arti cial intelligence,. The particularity of agent-based modelling is that it relies on the theoretical foundations of each domain and its conceptual vision of the world. This property offers exibility to each domain to represent in a reliable and realistic way the speci cities of the studied system. This session will focus on studies and feedback on
multi-agent modelling, arti cial population design and the use of machine learning for risk management in energy systems.

Leave a Reply

Your email address will not be published. Required fields are marked *