Unveiling the inter-relations between the urban streets network and its dynamic traffic flows

"The relation between micro-economic activity and space, like the relation between culture and space, is largely mediated by movement" [1].
Traffic flows have always been one of the major elements affecting the nature of urban streets. It influences the location of businesses, residence, the development of real estate, land values and built-density [2, 3]. While some scholars suggested that the morphology of the streets-network has the most significant effect on the traffic volume [4], others claimed that it results from the interactions of many urban attributes and agents [5]. One way or the other, revealing the relations between the static streets-network and the dynamic traffic-flows, may provide meaningful insights that could be used in planning processes. Thus, the objective of this work is to unveil the inter-relations between the urban streets-network and the dynamics of traffic. Percolation theory examines the robustness of networks to random attacks or failure. Here, we use network percolation analysis to identify street clusters that represent functional modules, composed of connected roads with traffic load lower than a defined threshold [6]. Based on the above, we identified spatially embedded clusters and tracked their dynamics at different temporal scales, ranging from one hour to a 7-day week. By doing so, we present the different patterns of urban mobility, its evolution, and dynamics.
We analyzed real data of the traffic flows in Tel Aviv and in London centers, which was collected from Google Directions API, every 15 minutes over a week (Tel Aviv: between 12-18/2/2017, and London: 21-27/3/2018). The data was collected for 5000-11500 streets segments that cover most of the examined areas. We analyzed the absolute velocities as well as their loads. While the analysis of the absolute velocities revealed the traffic patterns in the examined areas and pinpointed some anomalies (e.g., we identified Valentine day's eve), in this work, we focus on the analysis of the loads, which was divided into two parts. First, the dynamics of spatially embedded clusters, which allows us to follow the connectivity of specific locations to the rest of the area over time. Second, the dynamics of temporal clusters, i.e. clusters that were formed in specific hours throughout the entire week. In this analysis, we also studied the poorly connected areas in any examined time. Our findings reveal that the spatio-temporal clusters do not follow standard municipal definitions (e.g. neighborhoods) or correspond to specific land uses. This suggests that the city is better described by its dynamics rather than by its static structure alone. The importance of the methodology is that the information gained from the analysis can be used for various purposes such as land-use allocation, location-decision, real-estate value estimation, etc.

Authors: 
Nimrod Serok, Shlomo Havlin, Orr Levy and Efrat Blumenfeld-Lieberthal
Room: 
4
Date: 
Monday, September 24, 2018 - 17:00 to 17:15

Partners

The official Hotel of the Conference is
Makedonia Palace.

Conference Organiser: NBEvents

The official travel agency of the Conference is: Air Maritime

Photo of Thessaloniki seafront courtesy of Juli Bellou
fb flickr flickr

Contact

ccs2018@auth.gr