This paper is the second of two papers about the simulation of the traffic on the A9 Munich Autobahn in Germany; see ITRD E?????? for an abstract of the first paper, which covered data acquisition and the calibration of the FREQ model. The present paper covers the model's applications to traffic management applications. The calibration result provides close average speed predictions as compared with field-observed average speeds on a subsection and time interval basis. Selected studies were conducted to demonstrate the variety of traffic management strategies that can be studied with the FREQ model. The strategies analysed included: (1) the study of the impact of traffic incidents; (2) traffic growth and decline; (3) added auxiliary lanes; (4) ramp metering; and (5) added high-occupancy vehicle (HOV) lanes. A table summarises the results for all selected investigations, using base run results with no incident, with 0.95 growth factor, with 1.05 growth factor, with an auxiliary lane, with ramp metering, and with an added HOV lane. The emphasis of the A9 study was on the calibration of simulation modelling, the bridge between reality and modelling. It shows that only through careful model calibration can models be used to simulate traffic management strategies realistically.
Samenvatting