Route Choice of Cyclists in Zurich

摘要: This paper presents the first route choice model for bicyclists estimated from a large sample of GPS observations and overcomes limitations inherent in generally employed stated preference approach. It employs an improved mode detection algorithm post-processing to determine trips made by bicycle, which are map matched enriched street network. The alternatives generated as random exhaustive, but constrained search. Accounting similarity between with path-size factor MNL estimates show that elasticity regards trip length is nearly four times larger than respect share bike paths. product maximum gradient small. No other variable describing routes had impact. heterogeneity cyclists captured through interaction terms formulated on their average behaviour.