Document Type

Journal Article

Publication Date


Subject Area

infrastructure - vehicle, infrastructure - bus/tram priority, infrastructure - bus/tram lane, planning - methods, ridership - drivers, ridership - behaviour, mode - carpool


Violations, Video cameras, Vehicle occupancy, Validation, Statistical methods, Statistical analysis, Software validation, Seattle (Washington), Regulatory policy, Ramp metering, Ramp control, Queuing, Queues, Queueing, Questionnaires, Public opinion, Priority lanes, Policy, Policies, On ramps, Multiple occupancy vehicles, Motor vehicle operators, Mathematical statistics, Mathematical analysis, Logits, Logit models, Human behavior, HOV lanes, High occupancy vehicles, High occupancy vehicle lanes, Government policy, Field studies, Field data, Entrance ramps, Drivers, Diamond lanes, Design, Data collection, Data analysis, Data acquisition, Carpool lanes, Camcorders, Behaviour, Behavior


High-occupancy vehicle (HOV) lanes at metered on-ramps are intended to provide incentives for people to carpool or use transit. However, violations of the minimum passenger occupancy requirements penalize metered vehicles with extra delay and undermine the effectiveness and integrity of bypass lanes. Additionally, the perception of high violation rates can erode public support for HOV facilities. The purpose of this study was to gain a better understanding of the design factors and behavioral characteristics that influence violation rates at HOV bypass lanes at metered on-ramps. An on-ramp in Seattle, Washington was selected for data collection. This on-ramp consists of one metered lane and one HOV bypass lane. The site has a number of characteristics believed to influence violation behavior, including a minimum three-person vehicle occupancy requirement (most local facilities have only a two-person requirement), frequent extensive queuing, and little area for police enforcement. During data collection, all users of the HOV bypass lane were observed, and vehicle passenger occupancies and vehicle license plates were recorded. A high-speed video camera was utilized to assist with data verification. Subsampling of vehicle license plates from the metered lane was also performed. A mail-back survey with questions on travel habits, demographic and socioeconomic characteristics, and opinions toward and understanding of HOV and ramp metering policies was sent to the registered owner of each sampled vehicle. Ordered probit and binomial logit modeling, as well as general descriptive statistic methods, were utilized for survey analysis. Additionally, comparisons were made of subjects' observed behavior and their reported behavior. Findings from this study can help guide future policy on the implementation of HOV bypass lanes (and possibly other HOV facilities), identify appropriate methods to preserve the integrity of these facilities, and, in turn, maintain a positive public image.