FRANCISCO CHINESTA 

Arts et Metiers Institute of Technology (Paris)

Hybrid Artificial Intelligence for the Decision Making in Critical Urban Systems

BIO

Francisco Chinesta is Professor of computational physics at Arts et Metiers Institute of Technology (Paris), Fellow of the “Institut Universitaire de France” and of the Spanish Royal Academy of Engineering. He was and is involved in many industrial chairs AIRBUS, ESI, SKF, RTE and ESI-KEYSIGHT. He received many scientific awards: IACM Fellow, IACM Zienkiewicz, ESAFORM. He is author of more than 400 papers and more than 1200 conferences. He was president of the French association of computational mechanics and is at present president of the French Association of Mechanics, as well as director of the GdR I-GAIA on data and artificial intelligence based augmented engineering. He received many distinctions: Academic Palms, French Order of Merit, Doctorate Honoris Causa at the University of Zaragoza (Spain) and the Silver medal from the French CNRS.  He is the director of the DESCARTES project on Hybrid Artificial Intelligence for the Decision Making in Critical Urban Systems that the CNRS develops in its hub at Singapore.

 

ABSTRACT

Intelligent modelling technologies such as digital twins can enhance community-centric planning by simulating urban environments and developing predictive scenarios in response to critical and uncertain situations, allowing aided or even automated real-time decision-making. While traditional data-based AI techniques can be considered to enable the process, they still entail a number of limitations, such as the need of large amount of very specific and difficult-to-access data, massive computing resources, and questionable, sometimes risky output decisions in a few specific situations, thus lacking ethical considerations in decision-making affecting humans. The Hybrid Artificial Intelligence (HAI) approach aims to address these gaps. It combines AI's strengths of collecting and analysing big data across systems on an accessible online platform, with the pairing of physics-informed and physic-augmented approaches to allow for updating the knowledge. Technically, amodel of the ignorance (gap between knowledge-based prediction and the experimental observations) is constructed, and models are reduced by applying advanced parametric regressions techniques. This results in faster real-time diagnosis, prognosis and decision-making, using less data (frugality) yet with better accuracy in predictions, it remains almost explainable, and minimizes the ecological footprint (sustainability). By enabling human-driven updating of knowledge and security certifications, it also enables a human centric approach where the privacy of citizens is respected. Through HAI, different technological functionalities can be elaborated and, subsequently, closely combined for constituting the complex system of systems emulating the city functioning within its environment. This is applied to a number of case-based solutions for critical urban systems, where environmental maps (wind, pollution, temperature etc.), digital energy footprint, and smart sensing of large critical civil and industrial infrastructures are assimilated to inform efficient maintenance, optimize energy distribution, and enhance crisis and emergency management.