Effect of Link-Level Variations of Connected Vehicles (CV) Proportions on the Accuracy and Reliability of Travel Time Estimation

Abstract

The expected implementations of connected vehicles (CV) in the next few years will provide a promising alternative to existing technologies in providing data for travel time estimation. However, it is important to assess the quality of the estimation utilizing this new technology compared with existing technology. This quality is expected to be a function of the CV proportion in the traffic stream. This study proposes a method to estimate the CV proportion on specific network links in a region considering the variations in the socioeconomic characteristics between the zones within the region and a method to assess the impact of CV proportions on the quality of travel time estimation. The two methods are combined to determine the feasibility of using CV vehicles to estimate travel times for planning and operations purposes. Based on the application of the methods developed in this study, it can be stated that the CV proportion will be sufficient for use in planning and real-time operations of the investigated high demand freeway segment in the first year after the expected mandate for installing CV technology on all new vehicles will become effective. However, for the investigated high demand and high intersection density urban street, it will take one to three years for the data quality to be sufficient for use for planning purposes and three to six years for use for operation purposes, depending on the proportion of CV on the street segment, and considering the variations in the socio-economical characteristics in the region.