Real-time Multimodal Route Optimization and Anomaly Detection for Cross-border Logistics Using Deep Reinforcement Learning

Abstract

This paper presents a new approach to improve multiple choice and defect detection in cross-border shipments using deep learning (DRL). The design process involves the integration of real-time data from multiple sources to create comprehensive transportation models, including route optimization, cost reduction, and poor research methods. The DRL project is intended to use a multi-agent design to manage complex decision-making processes in a dynamic logistics environment. The hybrid anomaly detection system combines statistics with machine learning techniques to identify and respond to network disruptions. The system's performance was validated using a database including 185,432 shipment records collected over 24 months across the Asia-Pacific transportation system. The experimental results show that the proposed method has achieved 94.5% correct value in choosing the right path and 45% reduction in processing time compared to traditional methods. The negative detection antibody maintains a 96.2% true positive rate with a 1.8% false positive rate. The system's analysis shows that the growth of the needs in the calculation of the growth in the network, indicating the use of good resources in the large deployment. This research supports the state-of-the-art in cross-border business optimization by providing solutions that integrate real-time optimization methods with negative detection and response mechanisms.