Calibrating Simulation Models for Traffic during Incidents Using Shockwave Speeds Estimated from Field Data

Incidents have negative impacts on freeway operations. This study introduces a methodology for calibrating simulation models for freeway traffic during incidents. Queue length is used as a measure of effectiveness for calibrating a simulation model in VISSIM. Incidents of multiple durations were simulated and the resulting queue lengths were compared with observed queue lengths. The LWR model and shockwave speeds obtained from the field are used to estimate the observed queue lengths. The results show that the incident model in VISSIM can accurately represent the shockwave propagation speeds and queue lengths provided it is successfully calibrated. Such an incident model can be used to test alternative traffic management strategies to ameliorate the impact of incidents.