Prediksi Kemacetan Lalu Lintas Menggunakan Algoritma Random Forest

Abstract

Traffic congestion is a major issue in urban and suburban areas such as Pamulang, South Tangerang, caused by increasing vehicle volume without proportional road infrastructure development. This study aims to predict traffic congestion levels using the Random Forest algorithm and identify the dominant factors influencing congestion. The research utilized the Smart Traffic Management Dataset containing 2,000 observations with 12 variables such as average speed, traffic volume, weather conditions, and accident reports. The analysis process included data cleaning, categorical variable encoding, data splitting (80% training and 20% testing), and model validation using the 5-fold cross-validation technique. The results showed that the Random Forest model achieved 95,75% accuracy, with a precision of 0.96 and recall of 0.96. The feature importance analysis indicated that the average vehicle speed variable had the highest influence on congestion levels, followed by traffic volume and accident reports. These findings demonstrate that the Random Forest algorithm is highly effective in identifying traffic patterns and can serve as a foundation for developing intelligent transportation systems in the Pamulang area to reduce congestion adaptively and in real time.