CRASH PREDICTION ON ROAD SEGMENTS USING MACHINE LEARNING METHODS

This study adopted the Highway Safety Information System’s (HSIS) data for crashes occurred on road segments to develop supervised machine learning prediction models. Five models are developed: Linear Regression (LR), Generalize Additive Model (GAM), Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN). A comparison among five model was performed using root mean square error (RMSE) absolute (MAE) as quality indicators. The results indicated that RF found produce best crash results. findings suggested increase in Annual Average Daily Traffic (AADT) exponentially increased number of highway segments. In addition, roadway with higher design speed induced lower crashes, compared speed. For shorter than 5-mile long, rapidly segment length increased. However, there no substantial longer 5 miles. Also, greater lanes a segment, chance increasing crashes. Finally, moderate grades showed highest risk occurrences respectively followed by flat rolling grades. These useful transportation professionals consider when designing highways.