Statistical detection of anomalies in the economic behavior of traffic flows on road networks

In this work, we propose a methodology able to detect anomalies in economic behavior on a road network and apply it to the case of the Lazio region in Italy. First, we consider a complete statistical analysis of the road network. Then we analyze the traffic flows and the edge betweenness. Finally, using a K-Means algorithm on the traffic flows and the edge betweenness computed, we are able to identify the zones where we find significant discrepancies between the edge betweenness and the traffic flows. These anomalies can be ascribed to business activities or problems of the road networks and are possible sources of traffic over time.