<ul><li>Accurately estimating the impact of road maintenance schedules on traffic conditions is crucial to avoid excessive congestion during roadwork.</li><li>This paper explores the use of machine learning-based surrogate models to predict network-wide congestion caused by simultaneous road renovations.</li><li>XGBoost, among various regression models evaluated, stands out by significantly outperforming others in predicting traffic equilibria with a MAPE of 11%.</li><li>This approach has the potential to reduce the computational burden of large-scale traffic assignment problems in road renovation scheduling.</li></ul>

Machine Learning Predictions for Traffic Equilibria in Road Renovation Scheduling

Discover more