Calibration of VISSIM for Indian Heterogeneous Traffic Conditions

Abstract

Efficient modeling of vehicular traffic remains a largely debated issue especially in context of Indian heterogeneous driving conditions. VISSIM is a microscopic traffic simulation software that is gaining increasing recognition in India. However, modifications to the default behavioral parameters are essential to effectively simulate Indian heterogeneous traffic conditions. This paper presents a method and results on sensitivity analysis and automatic calibration of VISSIM model using data from an intersection in Chennai. This intersection has heavy flows during the peak time. VISSIM parameters affecting driving behaviour in Indian heterogeneous conditions were found using sensitivity analysis. ANOVA and elementary effects method were used in sensitivity analysis. The model was calibrated using Visual C++ COM interface of VISSIM. Genetic Algorithm was used to find the optimal combination of sensitive parameters during calibration.