Calibrating Route Choice Sets for an Urban Public Transport Network using Smart Card Data

Abstract

Identifying the set of alternatives from which travellers choose their routes is a crucial step in estimation and application of route choice models. These models are necessary for the prediction of network flows that are vital for the planning of public transport networks. However, choice set identification is typically difficult because while selected routes are observed, those considered are not. Approaches proposed in literature are not completely satisfactory, either lacking transferability across networks (observation-driven methods) or requiring strong assumptions regarding traveller behaviour (uncalibrated choice set generation methodologies (CSGM)). Therefore, this study proposes a constrained enumeration CSGM that applies the non-compensatory decision model, elimination-by-aspects, for choice set formation. Subjective assumptions of traveller preferences are avoided by calibrating the decision model using observed route choice behaviour from smart card data, which is becoming increasingly available in public transport systems around the world. The calibration procedure also returns two key insights regarding choice set formation behaviour: (i) the ranking of different attributes by their importance, and (ii) the acceptable detours for each attribute. To demonstrate the methodology and investigate choice set formation behaviour, the tram and bus networks of The Hague, Netherlands are used as a case study.

Understanding the Role of Cycling to Urban Transit Stations through a Simultaneous Access Mode and Station Choice Model

“…This study uses the calibration methodology proposed by Shelat et al. ( 23 ) (although slightly adjusted) to avoid having to assume behavioral parameters—that is, attribute ranking and thresholds—of the EBA model.…”

“…For the transit trip attributes, the general transit feed specification (GTFS) data associated with the network is used to generate different routes between stations using the same procedure as in Shelat et al. ( 23 ). Individuals are allowed an egress trip of less than 200 m between the destination station of the main trip and the observed destination station.…”

“…Transportation Research Record 2674 (8) strongly depend on the values of n or catchment radius selected by the researcher. The direct identification method proposed in the third category is also inherently unable to explain why certain alternatives were not considered in the dataset (23) and may suffer from endogeneity issues. Moreover, this method requires many observations per origin, which is typically not possible at the individual level and is thus usually applied at an aggregate level (5).…”

“…Although originally proposed as a probabilistic model (24), most choice set generation applications apply EBA as a deterministic model (22). This study uses the calibration methodology proposed by Shelat et al (23) (although slightly adjusted) to avoid having to assume behavioral parameters-that is, attribute ranking and thresholds-of the EBA model.…”

Selfish Yet Optimal Routing by Adjusting Perceived Traffic Information of Road Networks

“…In what follows, we use K i = 5. There are also several studies on generating the route candidates [63]- [65].…”