Public transport systems are generally organized in a static, a priori way. In such systems, the demand must be adapted to the offer. In this paper, we propose a model based on self-organization in order to dispatch a fleet of vehicles in a purely dynamic Transportation On Demand system (TOD). Our proposal consists in a decentralized approach and a multi-agent system (MAS) to model the environment. This will tackle the problem of vehicles over-concentration or the lack of service in certain areas of the city. We demonstrate that our model addresses these problems by providing vehicle agents, for a given request, to make the final decision thanks to a negotiation process and to calculate overcosts according to an original insertion heuristic.

Recent dramatic events recall to the world that is has to deal with risk problematic. Thus, to face risk in an agglomeration, we study hazards from natural or anthropic origin. One problem is to decide if it is necessary to evacuate or confine population. To help decision makers, we analyze the road network structure which may influence flow fluidity especially in a dangerous case. In this work, we detail an algorithm to detect communities in large graphs. It allows to identify routes that may cause problems in an evacuation case. Thanks to this algorithm, we study a toxic cloud propagation in a given zone and identify roads to avoid when evacuating this zone.

No abstract

Recently, the COVID-19 emerged in China and propagated around all the world has threatened millions of people and affected most countries and governments at several sides such as economical, educational, tourism, healthcare, etc. Indeed, one of the most important challenges that directly affect the people is the psychological side due to the harsh policies imposed by public authorities in most countries. In this paper, we propose a framework called CRISE that allows studying and understanding the psychological effect of COVID-19 during the lockdown period. Mainly, CRISE consists of four data stages: Collection, tRansformation, reductIon, and cluStEring. The first stage collects data from more than 2000 participants through a questionnaire containing attributes related to psychological effect before and during the lockdown. The second stage aims to preprocess the data before performing the study stage. The third stage proposes a model that finds the similarities among the attributes, based on the correlation matrix, to reduce its number. Finally, the fourth stage introduces a new version of Kmeans algorithm, called as Jaccard-based Kmeans (JKmeans), that allows to group participants having similar psychological situation in the same cluster for a later analysis. We show the effectiveness of CRISE in terms of clustering accuracy and understanding the psychological effect of COVID-19.