01025419

Component

Traffic signal timing plans are derived using deterministic tools operating macroscopically. We input link flows, turning proportions, and road network characteristics. The tools process the data and deliver signal timings with associated measures of effectiveness. This approach has served a generation of traffic engineers. However, we now have microscopic traffic simulation tools that embrace the stochastic nature of traffic. When we evaluate our deterministically derived plans in these stochastic simulators, we find that they sometimes deliver MOE's that are worse than those derived from signal timings before optimization. Our macroscopic optimizers are Synchro and TRANSYT-7F; our microsimulators are CORSIM, SimTraffic, and VISSIM. We also assess Direct CORSIM optimization, a feature of the new version of TRANSYT-7F. The nine-node grid network is the theoretical test bed modeling a wide range of traffic demands and distributions. Extensive experiments show that timing plans optimized by Synchro or TRANSYT7F can be subsequently evaluated by SimTraffic microsimulation. Direct CORSIM optimization delivers mixed results. Other combinations of macro-optimization and micro-evaluation provide inconsistent outcomes. The paper suggests that these inconsistencies need to be addressed through adjustment of the default simulation parameters.

01020180

06-1501

English

19p

2006

Transportation Research Board 85th Annual Meeting

CD-ROM

Figures (3) ; References (14) ; Tables (1)

Microsimulation; Optimization; Reliability; Traffic signal timing; Traffic simulation; Travel demand

CORSIM (Traffic simulation model); TRANSYT (Computer program)

Highways; Operations and Traffic Management; I71: Traffic Theory

Transportation Research Board Annual Meeting 2006 Paper #06-1501

BTRIS, TRIS, TRB

Mar 3 2006 10:42AM