Pre-calibration of a Discrete Choice Model and Evaluation of Cycling Mobility for Île-de-France

Guoxi Feng (1) , Maxime Jean (1) , Alexandre Chasse (1) , Sebastian Hörl (2)

1 IFPEN - IFP Energies nouvelles
2 IRT SystemX

An open and reproducible MATSim scenario of Île-de-France has recently been made available. In this paper, first, we aim to customize the discrete choice model for Île-de-France scenario and estimate its initial parameterization via the routing API OpenTripPlanner. The goal of this step is to reproduce the travel patterns observed from the regional Household Travel Survey (HTS) in MATSim without running any time-consuming simulation. Since the modeled cycling behavior could impact strongly the mode choice decision on bike, we evaluate the cycling routing with disaggregated GPS traces obtained dynamically from the mobile application Geovelo. The objective of this pre-calibration and evaluation is to provide a solid base for our future work which aims to calibrate the cycling flow per link in MATSim for Paris with the normalized flow from recent crowdsensing data.

Mots clés: Discrete Choice Model; pre-calibration; MATSim; GPS trace; cycling flow

Domaines: Modélisation et simulation; Analyse de données, Statistiques et Probabilités [physics.data-an]; Autre

Guoxi Feng, Maxime Jean, Alexandre Chasse, Sebastian Hörl. Pre-calibration of a Discrete Choice Model and Evaluation of Cycling Mobility for Île-de-France. 12th International Conference on Ambient Systems, Networks and Technologies (ANT), Mar 2021, Warsaw, Poland. pp.172-177, ⟨10.1016/j.procs.2021.04.003⟩. ⟨hal-03229508⟩