A novel congestion propagation modeling algorithm for smart cities

Abstract

Managing the frequent congestion of the road networks of large cities is a major challenge for city management organizations. In order to deal with these phenomena, it is essential to have an in-depth understanding of the processes that lead to the occurrence of congestion and its spilling over into contiguous areas of the city. The professional literature has a long history of studying the propagation of congestion, but so far the authors have focused on discovering frequent patterns and determining propagation probabilities. To the best of our knowledge, the Congestion Propagation Modeling Algorithm (CPMA) presented in this article is the first to be able to determine the expected value of propagation times for any propagation pattern as well as propagation probabilities using Markov chains. Our method was subjected to a detailed performance analysis using a dataset extracted from real road networks, during which we examined how accurately our model determines the propagation times of congestion, which is especially important for municipal traffic management organizations if they want to prevent the propagation of a local congestion and proliferation in the city. According to our measurements, our method can determine the probability of congestion occurring several times a day and the length of congestion propagation times with an accuracy that allows effective intervention by traffic management authorities.