A microsimulation calibration methodology based on speed-density relationships for freeways

Abstract

This paper describes the calibration methodology based on speed-density relationships in the microsimulation calibration process, stated that they represent the traffic flow phenomenon in a wide range of operational conditions and well summarize all the information that may be collected on field (or following a run of the microsimulation model) on two of the three key variables of traffic flow. The matching of speed-density relationships from field and simulation was evaluated using statistical analysis as technique of pattern recognition. A test freeway segment under uncongested traffic conditions was selected as case study; based on traffic data observed at A22 Brenner Freeway, Italy, statistical regressions between the variables of traffic flow were developed. Analogous relationships were obtained using the Aimsun microscopic traffic simulator software, reproducing the on field conditions and varying some selected parameters until a good matching between field and simulation was achieved