High-occupancy/toll lanes: where should they be implemented?

Abstract

Since the first high-occupancy toll (HOT) lane was implemented in Orange County, California in 1995 there has been an increasing interest in such lanes as a means to reduce congestion, increase utilization of underutilized high-occupancy vehicle (HOV) lanes, and generate revenue. In order to determine where such lanes are appropriate, a model is constructed to estimate how adding an HOV lane, an HOT lane or a mixed flow lane to an existing freeway affects delay in various circumstances. Except when the initial delay is very high, adding a mixed flow lane is more effective in reducing delay than adding either an HOV or HOT lane, and it costs less. If both the initial delay and the initial proportion of HOVs are very high, then adding an HOV lane is more effective in reducing delay than adding either a mixed flow or HOT lane. But if the initial delay is high and the initial proportion of HOVs is not high an HOT lane is the most effective in reducing delay. Adding an HOT lane should also be considered if substantial growth in traffic is expected, because HOT lanes perform relatively well in terms of reducing delay across all levels of initial delay and HOV usage. However, because HOT lanes are considerably more costly to construct and operate than mixed flow or HOV lanes, the revenue generating potential must also be assessed, and the savings in delay must be sufficient to justify the additional cost. The revenue generating capacity of HOT lanes is low when the initial delay is low, and in such cases alternate methods of utilizing the HOT lane initially might be desirable.