Transportation system management is a key to equilibrate the growing demand pattern with the limited transportation facilities. Various options of toll facilities provide an efficient solution to mitigate traffic congestion and meanwhile to increase the provider’s revenue. As a major source of provider’s revenue, several methods of toll collection are of interest to the transportation authorities to manage the toll demand by improving the convenience of electronic toll collection. What guarantees the accounted revenue and succeeds the transportation operation goals is the users’ willingness to pay toll cost in commuter-base real time traffic in contrast to travel time saving. This presentation analyzes different tolling scenarios to investigate price elasticity of travelers from various income levels, trip purposes, and vehicle groups. To reach this goal, a toll choice probability model, concentrating on the driver’s value of time, is developed to assign possible toll users to a transportation network. The toll choice model is embedded in between mode choice and traffic assignment steps of a statewide four-step travel demand model and is capable to forecast the change in toll demand by each income categories as a reason of pricing scenarios in the growing demand environment and real congested time. The approach is applied in Maryland Statewide Transportation Model (MSTM).

The choice probability of a toll road is estimated for each zone pair considering five household income groups of the origin zone. These spatial probabilities are differentiated by trip purpose and vehicle class. Toll demand is then derived from the auto trip matrices by applying these probabilities to each zone pair. In context of a travel demand model, both toll demand and non-toll demand proceed with four times of day estimated for each zone pair. Eventually in the traffic assignment step, potential toll users have the choice of toll road depending on their destination, with a penalty for non-toll users. The result of this study presents insight to travel behavior using toll choice model at a statewide level.