Random Sampling of Alternatives in a Route Choice Context
2007 (English)In: Proceedings of the European Transport Conference, 2007Conference paper (Refereed)
In this paper we present a new point of view on choice set generation and route choice modeling. Choice sets of paths need to be defined when modeling route choice behavior using random utility models. Existing approaches generate paths and assume that actual choice sets are found. On the contrary, we assume that actual choice sets are the sets of all paths connecting each origin-destination pair. These sets are however unknown and we propose a stochastic path generation algorithm that corresponds to an importance sampling approach. The path utilities should then be corrected according to the used sampling protocol in order to obtain unbiased parameter estimates. We derive such a sampling correction for the proposed algorithm. Furthermore, based on the assumption that actual choice sets contain all paths, we argue that Path Size (or Commonality Factor) attributes should be computed on all paths (or as many as possible) in order to reflect the true correlation structure. We present numerical results based on synthetic data. The results show that models including a sampling correction are remarkably better than the ones that do not. Moreover, unbiased estimation results are obtained if the Path Size attribute is computed based on all paths and not on generated choice sets. In real networks the set of all paths is unknown, we therefore study how many paths are needed for the Path Size computation in order to obtain unbiased results. The parameter estimates improve rather rapidly with the number of paths which is promising for real applications.
Place, publisher, year, edition, pages
Transport Systems and Logistics
IdentifiersURN: urn:nbn:se:kth:diva-48270OAI: oai:DiVA.org:kth-48270DiVA: diva2:457166
European Transport Conference 2007
QC 201111212011-11-172011-11-172011-11-21Bibliographically approved