Vehicle traffic is an important cause of pollution, especially in large metropolitan areas. Often the problem is caused by a non-suitable distribution of vehicles, which tend to follow shortest paths suggested by GPS navigators that do not take into account traffic or environmental conditions.

Recent Internet of Things (IoT) technologies and crowdsensing are today potential enablers to significantly extend the information spectrum for drivers and urban managers with the support of sophisticated monitoring systems able to exploit the large amount of data, coming from many sources, to provide “real time” high-level information.

The objective of the project is the implementation of an IoT and cloud based system to monitor the environmental conditions of road networks and to provide shortest paths from sources to destinations, taking into account both travel time and pollution. The project will be based on the adoption of technologies for distributed systems and big data processing. In particular, data will be organized according to graph-based models and the related metrics will be computed with paradigms for data parallelism.

The expected system is an extension of an infrastructure already designed and implemented for fast computation of betweenness centrality of weighted and directed graphs that model real road networks [1-6] and is part of a larger system for multi-modal mobility as a service.

The project will be also developed with the collaboration of the IFSTTAR research center of Lyon, the French Institute of Science and Technology for Transport, Spatial Planning, Development and Networks.

The expected system is an extension of an infrastructure already designed and implemented for fast computation of betweenness centrality of weighted and directed graphs that model real road networks [1-6] and is part of a larger system for multi-modal mobility as a service. An application of these services has been developed and presented in the context of the national competition "CINI Smart City University Challenge 2019" by a group of students from the Department of Engineering who ranked first (https://www.unisannio.it/en/articoli/university-sannio-wins-smart-city-university-challenge).

The student will be involved in the project activities and supervised by Prof. Eugenio Zimeo and Dr. Angelo Furno.

References:

[1] A. Furno, N.E. El Faouzi, R. Sharma, E. Zimeo: Fast Approximated Betweenness Centrality of Directed and Weighted Graphs. COMPLEX NETWORKS (1) 2018: 52-65.

[2] A. Castiello, G. Fucci, A. Furno, E. Zimeo: Scalability Analysis of Cluster-based Betweenness Computation in Large Weighted Graphs. IEEE BigData 2018: 4006-4015.

[3] A. Furno, N.E. El Faouzi, R. Sharma, V. Cammarota, E. Zimeo: A Graph-Based Framework for Real-Time Vulnerability Assessment of Road Networks. IEEE SMARTCOMP 2018: 234-241.

[4] Angelo Furno, Nour-Eddin El Faouzi, Rajesh Sharma, Eugenio Zimeo: Two-level clustering fast betweenness centrality computation for requirement-driven approximation. IEEE BigData 2017: 1289-1294.

[5] E. Henry, L. Bonnetain, A. Furno, N. E. El Faouzi, E. Zimeo: Spatio-temporal Correlations of Betweenness Centrality and Traffic Metrics. IEEE International Conference on Models and Technologies for Intelligent Transportation Systems, 2019.

[6] Antonio De Iasio, Angelo Furno, Lorenzo Goglia, and Eugenio Zimeo: A microservices platform for monitoring and analysis of IoT traffic data in smart sities. In Proceedings of the IEEE International Conference on Big Data, 2019.