Using state-of-the-art models in applied work: Travellers willingness to pay for a toll tunnel in Copenhagen. (December 2021)

• Record Type: Journal Article
• Authors: Lu, Hui; Hess, Stephane; Daly, Andrew; Rohr, Charlene; Patruni, Bhanu; Vuk, Goran
• Is Part Of: Transportation research. Volume 154(2021)
• Page Start: 37
• Page End: 52
• Publication Date: 2021-12
• Subjects: Value of travel time (VTT) -- Stated Choice (SC) experiment -- Multiplicative error structure -- Reference dependence -- Elasticities -- Random heterogeneity -- Correlation among alternatives -- Prototypical sampling; Transportation -- Research -- Periodicals; 388.011
• DOI: 10.1016/j.tra.2021.06.021
• Languages: English
• ISSNs: 0965-8564

Abstract

Highlights: Demonstration of effectiveness of advanced choice modelling in practical forecasting context. Evidence that the design of stated choice surveys impacts the resulting values of time (VTT). No significant difference in VTT observed between intercept and online panel respondents. Substantial random variation in VTTs for commute and other travel, in addition to socio-demographic and design effects. Importance of correlation between time period and route choice alternatives. Abstract: Advanced model specifications for value of travel time (VTT) research have become the norm in academic studies as well as large scale national and regional studies. However, studies with a quick turnaround and also those that need to produce results suitable for implementation in existing forecasting systems often still rely on simpler approaches. This paper describes how state-of-the-art specifications can benefit such studies. We focus on the willingness of car and light van travellers in the Copenhagen area to pay to use a proposed new route which includes the new Harbour Tunnel (Havnetunnel) and completes the Copenhagen Eastern Ring Road. We adopt the very general framework from the most recent UK VTT study (Hess et al, 2017 ), and extend it to capture the correlation among different alternatives in the choice presentations (which reflected both route choice and time of day choice). We find extensive heterogeneity across travellers, both deterministic and random. A sample enumeration procedure was then set up to calculate average VTT values for use in forecasting demand and appraising the new Ring Road investment.