Evaluation Of Operational Performance Between Different Types Of HOV Facilities In California: Continuous-Access Vs. Limited-Access

High-occupancy vehicle (HOV) facilities, which are reserved for vehicles with more than a pre-determined number of occupants, have been developed for decades as a part of the roadway network system to relieve pressures from ever-increasing travel demands and an effective and environmentally friendly approach to improve the mobility and productivity of freeway systems in metropolitan areas. Based on the traffic data from over 700 vehicle detector stations (VDS) during 2008 and 2009 period, a comparative study of operational performance at the route level was conducted between continuous-access and limited-access HOV facilities in California.The evaluation results of both HOV lanes and adjacent general purpose (GP) lanes revealed several operating characteristics of these lanes, including:1) the ingress/egress areas in limited-access HOV facilities may affect the formation of bottlenecks along HOV lanes; 2) the speed on HOV lanes and the speed differential between the HOV and adjacent GP lanes are statistically shown to be greater in continuous-access facilities than those in the limited-access facilities; and 3) the characteristics of speed-flow distribution of HOV lanes exhibit observable differences between the two types of HOV facilities, but those of adjacent GP lanes are similar regardless of the access type. Furthermore, statistical analyses show that some performance measures at the route level, including the space mean speed and vehicle-mile-traveled (VMT) share of the HOV lanes, are significantly different for HOV facilities with different access types.The accident-induced impacts were estimated by the proposed methodology, in which three major steps are included:1) synchronization of HOV lane accident information from two databases, the California Freeway Performance Measurement System (PeMS), and the Traffic Accident Surveillance and Analysis System (TASAS); 2) identification of accident impact regions; and 3) calculation of 10 accident-induced delays. The Wilcoxon rank sum test and log-linear regression models were used to investigate several influential factors on HOV lane accident impacts, such as lane configuration, accident duration, and injury or not. The analysis results imply that the impacts of HOV lane accidents on both HOV lanes and adjacent general purpose (AGP) lanes may be different for different HOV lane configuration, when compared the sample mean and variance of accident impact regions and delays, and also indicate that the accident duration and injury or not appear to be statistically significant (at 5% level) among all the factors examined in this study.