Managed lane operations are almost universally guided by a well-defined, overarching management strategy. Such a strategy might be aimed, for example, at achieving a target level of service. Managed lanes also seldom operate without competing obligations, such as a simultaneous commitment both to non-revenue users (e.g., buses, high occupancy vehicles) and to paying customers. Under contractual obligations to government partners, private operators face a bevy of unknowns or, at the very least, uncertainties surrounding the operations of the systems under their charge. In addition to those identifiable constraints and commitments, the performance of managed lanes is also subject to the unpredictable. Traffic patterns are continually changing and inherently random. While overall traffic patterns can be predicted or estimated, local changes in time and space can swing the operations of managed lanes for better or worse and can have downstream and compound effects.
TransModeler is a stochastic and probabilistic traffic model designed to capture the effects of these variables on managed lane systems and is capable of modeling traffic dynamics with high fidelity. With it, analysts can make offline planning predictions about how a system will perform under various management strategies, recurring traffic demand loads, and non-recurring incident scenarios. TransModeler is a high fidelity traffic simulation model that was designed to meet each of these criteria. This paper describes the software’s framework for managed lanes analysis and presents several examples of its application including SR 91 in Orange County, CA, and the Capital Beltway toll lanes in Northern Virginia.