Estimation d'un modèle agrégé de choix modal

Abstract : Aggregate modal choice model calibration. - Estimation of aggregated mode choice models has to face the problem of reliability of data used for calibration. Samples in travel surveys that are used for these estimations are generally too small for the needs of aggregate models. Therefore the number of observed trips per origin-destination is very small for the majority of them. Solution which is generally chosen needs on the one hand to reduce the number of zones by zones aggregation and on the other hand to keep for calibration only those origins-destinations for which sample size is above a minimum threshold. But we demonstrated that calibration results are highly sensitive to the choice of zones aggregation and to the choice of the threshold. In response to this problem this paper proposes an estimation method which allows very small-scale zoning to be retained and to keep the available information. The method is based on an iterative process whose convergence appears quite rapid and more importantly gives stable calibration results regarding zoning and threshold choices. To overcome one of the main problems of this method, which do not allow the calculation of confidence interval, we have used bootstrap method. This empirical analysis of mode choice between car and public transport on Lyon data allows us to study the benefits of our method in comparison with more conventional methods.
Document type :
Journal articles
Liste complète des métadonnées

Cited literature [13 references]  Display  Hide  Download
Contributor : Martine Sefsaf <>
Submitted on : Thursday, December 14, 2017 - 11:24:03 AM
Last modification on : Wednesday, October 31, 2018 - 12:24:05 PM


Explicit agreement for this submission

mis en ligne avec l'aimable autorisation de l'éditeur, AFITL


  • HAL Id : halshs-01020978, version 1



Patrick Bonnel. Estimation d'un modèle agrégé de choix modal. Les Cahiers scientifiques du transport , AFITL, 2013, pp.91-118. ⟨halshs-01020978⟩



Record views


Files downloads