Machine Learning-Enhanced Traffic Light Optimization System Prioritizing Emergency Vehicle Passage Using SVM and Random Forest Models

Traffic congestion in cities includes the complex and dangerous passing of emergency vehicles, which is a time-consuming task. This problem requires the optimisation of traffic lights in favour of emergency vehicles. To accomplish this, this paper discusses an optimized traffic light system using machine learning that prioritizes the passing of emergency vehicles into city areas. It integrates SVM and Random Forest models by dynamically adjusting traffic light signals based on traffic density to accelerate emergency vehicles. The results reveal that the proposed system would lead to improved emergency response times while enhancing overall transportation efficiency with reduced congestion of traffic. Additionally, the study further went on to establish the effectiveness of the proposed model as a solution in traffic flow optimization and management. Results show that the performance of the proposed model is effective for the purpose of traffic light optimization. The SVM+SAFS and RF+SAFS methods figured prominently as high-performance methods with accuracy rates of 94.89% and 95.02%, respectively. Furthermore, in the case of the RF+SAFS method used for traffic light optimization, it was possible to reduce the average waiting time by 20%, increase the capacity of transit by 15%, and decrease fuel consumption by 10%. Overall, combining the outputs in the model led to the following performance, an 18% decrease in total travel time.