To What Extent Can High-Occupancy Vehicle Lanes Reduce Vehicle Trips and Congestion?

Abstract

Despite the increasing interest in and deployment of managed lanes on highways that involve pricing and tolling schemes, a considerable number of highway facilities still use high-occupancy vehicle (HOV) lanes with no other features to manage or price their use. These HOV lanes are often restricted to vehicles transporting two or more, or three or more, occupants and are viewed as mechanisms to reduce vehicular travel demand, motivate ridesharing among individuals, and curb greenhouse gas emissions. In this paper, an analysis of the 2001 National Household Travel Survey data set was conducted to determine the potential share of HOV trips that might be induced by the presence of an HOV lane. It was found that a vast majority of HOV trips were trips undertaken with family for discretionary activity purposes. Given that virtually all these HOV trips would have been undertaken regardless of the presence of an HOV lane, one could question the potential efficacy of implementing a pure HOV lane. This paper addresses this question by considering two extreme scenarios that provide a range of potential impacts of HOV lanes on vehicle trip reduction. In the specific hypothetical scenario considered in this paper, the range was found to lie between 1% and 37%, with a clear indication that true impacts in the real world were more likely to be closer to the lower bound than the upper bound. On the basis of these findings, it appears that pure HOV lanes provide modest benefits from the perspectives of vehicular trip reduction and congestion relief. The paper concludes that HOV lanes with additional managed use and pricing features are likely to yield substantially greater impacts from the perspectives of both vehicular trip reduction and revenue generation.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.