IoT-Assisted Autonomous Urban Delivery Vehicles using Random Forest for Intelligent Logistics Planning

The rising need for effective last-mile delivery has prompted the creation of Internet of Things (IoT)-enabled Autonomous Urban Delivery Vehicles (AUDVs) capable of navigating and functioning intelligently in intricate urban settings. This research presents an innovative logistics planning system that combines real-time IoT data with a Random Forest algorithm to improve delivery efficiency, route optimization, and operational decision-making. This objective is to develop intelligent logistics planning systems that use IoT data and machine learning to enhance delivery routes, save time and fuel consumption, and increase dependability. The system uses sensor data, including traffic density, delivery priority, vehicle load, and weather conditions, to forecast ideal delivery routes and timetables. Experimental validation was performed with a simulated urban logistics dataset consisting of 10,000 delivery records under diverse traffic and weather conditions. The proposed Random Forest-based planning model realized a 21% decrease in average delivery time, a 17.8% enhancement in fuel economy, and a 19.6% rise in successful on-time deliveries, relative to a baseline rule-based routing system. The program achieved a prediction accuracy of 94.6% in identifying ideal delivery routes. The findings illustrate the efficacy of combining machine learning with IoT in autonomous logistics systems, providing a scalable solution for transport in smart cities.