Citation

Saadi I, Mustafa A, Teller J, Cools M. Transp. Policy 2018; 67: 102-110.

Copyright

(Copyright © 2018, Elsevier Publishing)

DOI

10.1016/j.tranpol.2017.09.009

PMID

unavailable

Abstract

In Belgium, river floods are among the most frequent natural disasters and they may have important consequences on travel demand. In order to better understand how the travel patterns vary, we propose to set up a large scale scenario based on MATSim for guarantying an accurate assessment of the impact of river floods on the transportation system. As inputs, the current agent-based model requires a base year population. A synthetic population with respective set of attributes is generated as a key input. Afterwards, agents are assigned activity chains through an activity-based generation process. Finally, the synthetic population and the transportation network are integrated into MATSim. Regarding data, households travel surveys, OD matrix of Belgium have been used to set up the demand. For simulating river floods effects, a steady-state inundation map has been integrated within MATSim. In the current study, five scenarios have been tested where critical links are associated various levels of service, i.e. 10%, 25%, 50%, 75% and 100% (base case scenario). They are systematically compared to the standard scenario to estimate the deviations in terms of traffic patterns and travel times. The results suggest that compared to the standard scenario, the average trip travel time increased by 16.36%, 44.44%, 126.77% and 144.44% with respect to scenarios 75%, 50%, 25% and 10% respectively. Also, the traffic flows have been re-distributed more uniformly across the transportation network. Roads with important traffic volumes are subjected to a decrease of activity on the contrary of roads with low traffic volumes. A very few studies have focused on how river floods affect transportation systems, this paper provided new insights in term of methodology and traffic patterns analysis under disruptions.

Language: en

Keywords

Agent-based modeling; MATSim; River floods; Travel demand