Comparing Calibrated Shared Space Simulation Model with Real-Life Data

Shared spaces are being implemented in many countries to deal with safety concerns and traffic flow problems on busy urban streets and street crossings. However, shared space concepts could not be tested before they were built because of the lack of a functioning microscopic shared space simulation. Lane-based car-following models, currently used in traffic simulation, cannot reproduce the high heterogeneity of a mixed traffic mode's nonchannelized flow. This paper's novel approach introduces an extended social force model for vehicles and pedestrians that incorporates social interactions between different modes of transport rather than following a purely rule-based approach. The calibration of such a microscopic traffic simulation model with real-world data from two shared space sites is presented. The simulation can reproduce real-life shared space behavior by comparing it with trajectory and interaction data collected at implemented shared space road designs.