Assessment of TexSIM capabilities for analyzing freeway ramp metering

Abstract

As traffic volumes continue to increase in urban and suburban areas and there is less space to increase freeway capacity by means of lane additions, freeway traffic operations must be improved using other means by adopting new control strategies. Since freeway simulation plays a vital role in arriving at efficient and effective freeway ramp control strategies, a new traffic simulation model that is being developed at Texas A&M University, called TEXSIM, will be assessed for its simulation capabilities. The purposes of this study were to calibrate the TEXSIM model for various traffic facilities and to assess its capabilities of freeway simulation with various ramp metering strategies. Due to lack of suitable field data, the TEXSIM model was first calibrated and compared to other field verified models such as PASSER 11-90, PASSER 111-90, NETSIM and FRESIM. This testing would assess the robustness of the model for its queue dynamics, stop-and-go characteristics at a traffic signal, lane changing capabilities, etc. Several traffic facilities such as an isolated intersection, a diamond interchange and a freeway segment with and without ramp control were used in the comparisons. A series of scenarios covering a range of traffic volumes and operational features were tested in the simulations. The simulation results indicate that for pretimed isolated intersections, TEXSIM compares well against PASSER 11-90 and NETSIM. Although a statistically significant difference was found between PASSER 11-90 and TEXSIM at the ninety-five percent confidence level, the simulation results were observed to be comparable. The same hypothesis was not rejected for the data set having only through movement delays. For diamond interchanges, the statistical analysis showed that there is no significant difference between TEXSIM and NETSIM at the ninety-five percent confidence level. The results also showed that all three programs (PASSER 11-90, NETSIM and TEXSIM) are close to one another. The comparison of speed-flow-density relationship curves produced by TEXSIM with those generated by FRESIM showed that TEXSIM works better for freeway simulation due to its consistent results with respect to the input parameters. It was also observed that for simple clock time ramp metering, the delay curve of TEXSIM is very close to that of FRESIM and the delay curves of TEXSIM and FRESIM compare closely to the HAM delay curve The study also concluded that dual-lane metering provides various advantages over traditional single-lane metering including more ramp storage capacity, reduction of queue lengths, etc., thereby reducing the queue spill-back on the frontage roads. It was also recommended that further research be carried out to study various ramp metering strategies using an actuated ramp control.