A Queueing Theory Approach to Modeling Toll Plaza Delay with Applications for Commute User Cost Optimization

Abstract

Queues at toll plazas are often the source of long delays for commuters trying to get from their residence to workplace. Commuters traveling from New Jersey to Manhattan can choose from three tollways to cross the Hudson River - the Holland Tunnel, Lincoln Tunnel, and George Washington Bridge. The optimal route, however, for the same origin in New Jersey and destination in Manhattan may not always be the same because of changing delay times at toll plazas throughout the day. This thesis focuses on the minimization of user cost, which includes travel time and vehicle operating costs, to solve the optimal commute problem. Empirical hourly traffic volume data and queueing theory are used to model delay time and create delay pro les for each crossing. We estimate the price elasticity of travel demand to find a relationship between toll prices charged and the delay profile. Finally, we develop a a user application that allows drivers to determine their optimal route by inputting factors including their origin, destination, and time of travel.