Analysis and Classification of the Vehicular Traffic Distribution in an Urban Area Jorge Luis Zambrano-Martinez 1(B ) , Carlos T. Calafate 1(B ) , David Soler 2 , Juan-Carlos Cano 1 , and Pietro Manzoni 1 1 Department of Computer Engineering (DISCA), Universitat Polit` ecnica de Val` encia, Valencia, Spain jorzamma@doctor.upv.es, {calafate,jucano,pmanzoni}@disca.upv.es 2 Institute of Multidisciplinary Mathematics (IMM), Universitat Polit` ecnica de Val` encia, Valencia, Spain dsoler@mat.upv.es Abstract. Nowadays, one of the main challenges faced in large metropolitan areas is traffic congestion. To address this problem, an ade- quate traffic control could produce many benefits, including reduced pol- lutant emissions and reduced travel times. If it were possible to charac- terize the state of traffic by predicting traffic conditions, measures could be taken to preventively mitigate the effects of congestion and related problems. This paper performs an experimental study of the traffic dis- tribution in the city of Valencia, characterizing the different streets of the city in terms of vehicle load with respect to the travel time during rush hour traffic conditions. Experimental results based on realistic vehicu- lar traffic traces show that most of the street segments under analysis present a good fit under quadratic regression, although a large number of street segments fall under other categories mainly due to lack of traf- fic. Based on this study, a clustering analysis study associated to the different streets shows how these streets can be classified into four inde- pendent categories, evidencing an uneven traffic distribution throughout the city. Keywords: Traffic prediction · Traffic behavior · Clustering · Urban traffic · Valencia 1 Introduction Traffic congestion can significantly increase the travel time of vehicles, being directly associated to increased delays, an inefficient use of fuel, and an increase of CO 2 emissions, being all of them critical issues for city authorities [1]. As we gradually move towards a new paradigm centered on automated vehicles, we are able to empower traffic administrators with more sophisticated ways of regulating traffic compared to the usual strategies based on semaphore timing regulations, or the deployment of traffic agents on site. Among these novel tech- niques of handling traffic, centralized route management emerges as a solution c  Springer International Publishing AG 2017 A. Puliafito et al. (Eds.): ADHOC-NOW 2017, LNCS 10517, pp. 121–134, 2017. DOI: 10.1007/978-3-319-67910-5 10