Impact of different car-following models on estimating safety and emissions on signal-controlled intersections using microscopic simulations

This paper examines the influence of two selected car-following models on the outcomes of microscopic traffic simulations. The authors begin by reviewing the literature on the various traffic models, methods for estimating energy consumption, fuel use, and emissions. The authors discuss using surrogate safety measures derived from analysing vehicle trajectories in a microscopic traffic model to estimate safety levels. This paper also outlines the authors' approach to data acquisition and processing at the chosen test site. Most data are sourced from the city's Intelligent Transport System (ITS) services, including traffic flow volume, vehicle speed, and public transport travel times. The data gathered from Automatic Vehicle Location (AVL) systems was compared to manual travel time measurements. Depending on the analysed section Mean Average Error (MAE) ranging from 3 to 5 seconds was obtained, however, Mean Absolute Percentage Error (MAPE) ranged from 7.46% to 30.01%. The proposed method aims to evaluate the precision of the microscopic model in replicating real road networks and traffic based on available datasets. Two car-following models (CFM), Wiedemann 74 (W74), and Wiedemann 99 (W99) are chosen for further analysis. The models were validated using the GEH statistic and by comparing speed distributions. Both the models yielded satisfactory results. Different scenarios involving changes in traffic flow and to traffic signal configurations were analysed. While both W74 and W99 yielded satisfactory results, there were discernible disparities in the analyses. In general, W99 yields less favourable results, characterised by emissions, and delays compared to W74. Depending on the scenarios compared, the number of conflicts for W74 varied up to 56% (scenarios 2 and 4), and up to 78% for W99 (scenarios 1 and 4). The methodology presented by the authors can be used in future research to analyse various traffic control solutions concerning their impact on the environment and traffic safety. The observed differences underscore the importance of careful model selection and presented approach can serve as a foundation for developing guidelines for microscopic modelling and a multi-criteria approach to selecting the most effective implementation scenarios.