Improved Traffic Congestion Level Indicator: A Fuzzy Logic-based Approach

Traffic congestion is a prevalent urban phenomenon. Extensive research has been conducted to support the mitigation of traffic congestion. However, an acceptable definition of traffic congestion levels is still elusive, and it generally relies on uncertain terminology to describe the severity. Therefore, this research proposes a conceptual model based on fuzzy logic to quantify and interpret traffic congestion levels. Parameters such as the number of vehicles, the number of lanes on the road, the percentage of different vehicle types passing through the area, the area occupied on the road by vehicles, and the time period under consideration have been incorporated into the model, taking into account similar research and identified gaps. A methodology has been devised to provide a numerical representation of real-time traffic congestion. A fuzzy inference system has been introduced to interpret traffic congestion using uncertain terms. The validity of the proposed system has been assessed using a confusion matrix. Furthermore, Cohen's Kappa analysis was employed to evaluate the overall accuracy of the system, revealing an observed Kappa value of 0.82 (> 0.8), signifying a high level of accuracy. This model holds the potential to benefit commuters and policymakers in developing more sustainable and resilient traffic management systems.