Urban travel time reliability analysis with consumer GPS data

This study explored the use of commercially available consumer GPS data in travel time reliability analyses. Specifically, speed profiles provided by TomTom, Inc., were used to obtain travel time statistics (i.e., means and variances) on nonexpressway road segments in the Chicago, Illinois, metropolitan area. Travel time distributions were then derived for these road segments to complement those obtained from conventional expressway traffic sensor data. The analysis of travel time reliability was based on the theory of stochastic dominance, which attempted to capture the commonality of all individuals who had similar risk-taking preferences. A case study built on this new data source was used to examine (a) how risk-taking preferences (risk neutrality, risk aversion, and ruin aversion) would lead to different admissible path sets and (b) how the new data sources would affect the results of travel time reliability analysis.