Quantifying the Attractiveness of High-Occupancy Toll Lanes with Traffic Sensor Data under Various Traffic Conditions

Different states have used different tolling strategies for their high-occupancy-toll (HOT) lane facilities, and the performance of an HOT lane system has tied directly to the tolling strategy. However, comparing tolling strategies has been a difficult problem because of the lack of field data as well as missing pieces of information critical to the analysis of HOT lane operations. For example, how single-occupancy vehicles (SOVs) choose the type of lane to use under the effect of tolling and the flow friction between HOT and general purpose lanes remains unclear. This research attempts to analyze the heterogeneity among the SOV users of HOT lanes and to quantify the attractiveness of HOT lanes for SOV drivers. The study uses field data collected by point traffic sensors and transponder toll tags on Washington State Route 167 (SR-167). SOVs are categorized as infrequent users and frequent users, according to the frequency with which their transponder identification tags appear in the database. Their value of time distributions are studied separately at the microscopic level. A logit-like model is further developed to quantify the relationship between SOV demand and HOT lane use under different traffic phases. The analysis results show that travel patterns on HOT and general purpose lanes can be better interpreted and modeled for performance enhancement with motorists' willingness to pay HOT lane tolls considered.