Evaluation of Freeway Traffic Data Acquisition: Technology, Quality, and Cost

Abstract

This study evaluates the accuracy of the travel time data that are estimated by Dual loop, Waze, HERE, and INRIX against Bluetooth data on multiple segments of the I-80 freeway between Davis and Sacramento, CA. We conduct a simulation-based critical sampling rate analysis, which suggests that the Bluetooth travel time data qualifies for approximating the ground truth. Further, we apply evaluation methods and derived indices, Travel Time Error Bias (TEB), Average Absolute Travel Time Error (ATE), and Standard Error of Mean (SEM), to compare the travel time data reported by Waze, HERE, and INRIX with the benchmark (Bluetooth) data on the selected testing segments on I-80. The results show that the INRIX and HERE data closely match the Bluetooth data, both in the trends and values of reported travel time. Moreover, all three vendors’ data accuracy deteriorates when the traffic congestion intensifies.