A strategic double-ended queuing model for shared transportation

Abstract HOV lanes are restricted to vehicles with more than one person, and ride sharing can be used to be eligible to use them. This paper provides an analysis of such a system from the perspective of strategic queuing. Our model is unique with respect to the queuing theory literature, in that one who joins the corresponding double-ended queue chooses between doing it as a customer or as a server. We also allow travelers to choose unrestricted lanes. We study equilibrium strategies that agents might follow and the optimal strategies from a social view point and compare the two. The model is studied under two information levels, with the goal of determining the optimal number of passengers required for the HOV lanes.