Evaluation of freeway control using a microscopic simulation laboratory

摘要

This paper presents a microscopic simulation laboratory, microscopic traffic flow simulator (MIT-SIM)Lab, and its application to evaluation of freeway traffic control. Traffic pattern on roadways is emergent from a complex interaction between driving behavior and control systems. The uniqueness of MITSIMLab lies in a realistic representation of this interaction. Various components of MITSlMLab are presented. Evaluation is performed for the Central Artery/Tunnel network in Boston. Four kinds of traffic control are evaluated: access, route, lane and integrated control. Performance of various traffic control strategies is investigated under a range of traffic conditions and parametric values governing their operations. Results from this study substantiate findings from previous field and simulation evaluations of ramp control and provide new insights with regard to the effect of queue override strategy, downstream traffic condition and their interactions on the performance of ramp metering. Lane control evaluation suggests the need for a case-by-case study, whereas route control study demonstrates the deficiency of a reactive route guidance system. Finally, it is shown that a systematic design for integrated traffic control needs to be developed in order to exploit the potential benefit.