Understanding the Behavior of Travelers Using Managed Lanes - A Study Using Stated Preference and Revealed Preference Data

This research examined if travelers are paying for travel on managed lanes (MLs) as they indicated that they would in a 2008 survey. The other objectives of this research included estimating travelers? value of travel time savings (VTTS) and their value of travel time reliability (VOR), and examining the multiple survey designs used in a 2008 survey to identify which survey design better predicted ML traveler behavior.

To achieve the objectives, an Internet-based follow-up stated preference (SP) survey of Houston?s Katy Freeway travelers was conducted in 2010. Three survey design methodologies?Db-efficient, random level generation, and adaptive random?were tested in this survey. A total of 3,325 responses were gathered from the survey, and of those, 869 responses were from those who likely also responded to the previous 2008 survey.

Mixed logit models were developed for those 869 previous survey respondents to estimate and compare the VTTS to the 2008 survey estimates. It was found that the 2008 survey estimates of the VTTS were very close to the 2010 survey estimates.

In addition, separate mixed logit models were developed from the responses obtained from the three different design strategies in the 2010 survey. The implied mean VTTS varied across the design-specific models. Only the Db-efficient design was able to estimate a VOR. Based on this and several other metrics, the Db-efficient design outperformed the other designs. A mixed logit model including all the responses from all three designs was also developed; the implied mean VTTS was estimated as 65 percent ($22/hr) of the mean hourly wage rate, and the implied mean VOR was estimated as 108 percent ($37/hr) of the mean hourly wage rate.

Data on actual usage of the MLs were also collected. Based on actual usage, the average VTTS was calculated as $51/hr. However, the $51/hr travelers are paying likely also includes the value travelers place on travel time reliability of the MLs. The total (VTTS+VOR) amount estimated from the all-inclusive model from the survey was $59/hr, which is close to the value estimated from the actual usage. The Db-efficient design estimated this total as $50/hr.

This research also shows that travelers have a difficulty in estimating the time they save while using a ML. They greatly overestimate the amount of time saved. It may well be that even though travelers are saving a small amount of time they value that time savings (and avoiding congestion) much higher ? possibly similar to their amount of perceived travel time savings.

The initial findings from this study, reported here, are consistent with the hypothesis that travelers are paying for their travel on MLs, much as they said that they would in our previous survey. This supports the use of data on intended behavior in policy analysis.