Revealing the day-to-day regularity of urban congestion patterns with 3D speed maps

Clélia Lopez,Ludovic Leclercq,Hans Van Lint,Panchamy Krishnakumari,Nicolas Chiabaut

doi:10.1038/s41598-017-14237-8

Abstract

In this paper, we investigate the day-to-day regularity of urban congestion patterns. We first partition link speed data every 10 min into 3D clusters that propose a parsimonious sketch of the congestion pulse. We then gather days with similar patterns and use consensus clustering methods to produce a unique global pattern that fits multiple days, uncovering the day-to-day regularity. We show that the network of Amsterdam over 35 days can be synthesized into only 4 consensual 3D speed maps with 9 clusters. This paves the way for a cutting-edge systematic method for travel time predictions in cities. By matching the current observation to historical consensual 3D speed maps, we design an efficient real-time method that successfully predicts 84% trips travel times with an error margin below 25%. The new concept of consensual 3D speed maps allows us to extract the essence out of large amounts of link speed observations and as a result reveals a global and previously mostly hidden picture of traffic dynamics at the whole city scale, which may be more regular and predictable than expected.

Highlights

In this paper, we aim to pursue the investigation of regularity in macroscopic mobility patterns not by focusing on the commuting flow distributions; but on the resulting level of service of the transportation network, i.e. on congestion patterns
It should be noticed that all the methods elaborated in this paper can be applied to any set of time-dependent link speed data combined with the related connected graph whatever the initial sensing method is
The global analysis of Amsterdam link speed data over 35 days shows a high degree of regularity when comparing the daily congestion patterns

Summary

Introduction

We aim to pursue the investigation of regularity in macroscopic mobility patterns not by focusing on the commuting flow distributions; but on the resulting level of service of the transportation (road) network, i.e. on congestion patterns. We synthesize within days link speed data and simplify day-to-day comparisons, by means of so-called spatio-temporal speed cluster maps Such 3D speed maps consist of a joined partition of space (road network links) and time (the different observations) into homogeneous clusters characterized by a constant mean speed. We will show that using a single consensus pattern for each class of 3D congestion maps is sufficient to accurately estimate in real-time travel times in the city This means that addressing congestion patterns directly at the whole city scale for all time intervals reveals a meaningful and accurate global picture of the city traffic dynamics that can be used as an efficient alternative to classical methods that process much more data at local and short-term scales

Objectives

Methods

Results

Conclusion