The value of travel time and reliability-evidence from a stated preference survey and actual usage

This research examined travel behavior of Managed Lane (ML) users to better understand the value travelers place on travel time savings and travel time reliability. We also highlight the importance of survey design techniques. These objectives were accomplished through a stated preference survey of Houston's Katy Freeway travelers. Three stated choice experiment survey design techniques were tested in this study: Bayesian (D-b) efficient, random level attribute generation, and an adaptive random approach. Mixed logit models were developed from responses using each of those designs. The value of travel time savings (VTTSs) estimates do vary across the design strategies, with the VTTS estimates based on the D-b-efficient design being approximately half the estimates from the other two designs. However, among the three design strategies, the value of travel time reliability (VOR) was only significant in the D-b-efficient design. The estimated VTTS from actual Katy Freeway usage (as measured using actual tolls paid and travel time saved on the managed lanes) is $51/h. This likely also includes any value that travelers place on travel time reliability. In comparison, our combined estimate of VTTS and VOR based on the SP survey (D-b-efficient design) was $50/h, which is remarkably close to the estimate from the actual usage data. Based on our dataset, the D-b-efficient design technique proved superior to the other two techniques. Finally, this research also supports the importance of including both travel time and travel time reliability parameters when estimating the willingness to pay for, and therefore benefits derived from, ML travel. (C) 2012 Elsevier Ltd. All rights reserved.