Application of Dynamic Value Pricing through Enhancements to TRANSIMS

In the past decade, transportation agencies have become increasingly interested in a high-occupancy toll (HOT) lane value pricing system, especially one with a dynamically varying toll system that depends on the congestion levels in the HOT lanes. HOT lane users are influenced by many factors, including toll prices, savings in time, and purpose of the trip. The current methodologies that forecast the demand for a HOT lane value pricing system are aggregate and do not consider individual traveler socioeconomic characteristics, particularly the value of time (VOT). This paper applies heterogeneous VOT for each individual into the generalized travel time function. This approach relaxes the conventional assumption of constant VOT. The generalized travel time function combined with VOT is used as a route choice for each traveler between the two choices (a toll lane versus a nontoll lane) in a HOT lane value pricing system. With this function, each driver is assigned to the network on the basis of a dynamic traffic assignment. This approach is implemented by using TRANSIMS on a large transportation network. TRANSIMS has a microscopic simulation model that enables the user to capture route choice behavior, including each traveler's response to dynamic tolls through heterogeneous VOT. To initiate this capability, TRANSIMS was enhanced to support dynamic toll pricing. It is now capable of presenting the 15-min dynamic toll rates. In addition, the impacts of various tolls on route choice can be analyzed on the basis of socioeconomic and trip characteristics of each traveler.