A Calibration and Validation Procedure for Microscopic Simulation Model: A Case Study of SimTraffic for Arterial Streets

Microscopic simulation models are becoming increasingly important tools in modeling transport systems. The main reason is that simulation is faster, safer, and less expensive than field implementation and testing. While these simulation models can be beneficial, the models must be calibrated and validated before they can be used to provide meaningful results. This paper proposes a calibration and validation procedure for microscopic simulation model. While previous studies focused mostly on model calibration to study a freeway section, this procedure focuses on model calibration to study an arterial segment that includes signalized intersections. The procedure was applied on the microscopic simulation model SimTraffic on a real world traffic data from two sites in Florida to calibrate a model to study weaving movements along arterial streets. The data from the first site was used in the calibration process and the data from the second site was used in the validation process. The proposed procedure consisted of the eight steps: i) identification of measures of effectiveness; ii) data collection; iii) identification of calibration parameters; iv) determination of number of simulation runs per scenario; v) determination of total number of simulation runs; vi) visualization of the animation; vii) relative error; and viii) validation with a new data. The proposed procedure appears to be properly calibrating and validating the SimTraffic simulation model for the two test sites. TRB 2005 Annual Meeting CD-ROM Paper revised from original submittal.