Advanced Modeling and Interpretation for Accurate Intersection Traffic Time Prediction

This study presents the results for predicting waiting time of vehicles at intersections. Various traditional models (LR, DT, RF, GB, KNN, MLP, SVM) and H2O-SEM approach are compared based on Mean Square Error (MSE), Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and coefficient of determination. Additionally, the H2O AutoML module is introduced as an alternative for comparison. The traffic generation and investigation are performed using the SUMO simulator. The findings aim to identify the most effective model for accurate waiting time prediction at intersections.