Optimizing Logistics Transportation Efficiency in India: An Enhanced Predictive Model Using Machine Learning Techniques

The rapid advancement of the logistics and transportation sectors has underscored the critical importance of efficient travel management and the punctuality of transportation vehicles. This study undertakes a rigorous analysis of logistics data in India, aiming to enhance the efficiency of logistics transportation. A comprehensive predictive model was developed, integrating multiple influencing factors and involving a comparative optimization of models including Logistic Regression, Decision Tree, and Random Forest. Utilizing the Delivery Truck Trip Data from 2019 to 2020 on the Kaggle platform, extensive data preprocessing, novel variable computation, and feature engineering were conducted. Results indicate that, post parameter tuning, the Random Forest model exhibited superior performance in delay prediction for drivers. Feature importance analysis revealed that GPS provider, transportation distance, expected trip time, and the month were significantly influential in predicting delays. This research not only augments the theoretical framework of delay prediction but also offers practical solutions for logistics enterprises to enhance operational efficiency and costeffectiveness.