🤖 AI Summary
To address the significant deviation between theoretically optimal routes and actual execution in last-mile delivery—caused by drivers dynamically adjusting stop sequences based on experience—this paper proposes a practical stop-sequence prediction method leveraging historical trajectory data. Our approach introduces a novel pairwise-local attention mechanism tailored to stop pairs and integrates an iterative first-stop selection algorithm, jointly modeling local pairwise dependencies and optimizing global operational cost. The model adopts an Encoder-Decoder architecture enhanced with a Pointer Network to enable end-to-end sequence generation. Evaluated on a real-world Amazon US dataset, our method improves top-4 stop prediction accuracy from 0.20 to 0.312 and reduces sequence-level discrepancy between predicted and ground-truth routes by approximately 15%. These results demonstrate substantial improvements in routing plan practicality and deployability for real-world last-mile logistics operations.
📝 Abstract
In last-mile delivery, drivers frequently deviate from planned delivery routes because of their tacit knowledge of the road and curbside infrastructure, customer availability, and other characteristics of the respective service areas. Hence, the actual stop sequences chosen by an experienced human driver may be potentially preferable to the theoretical shortest-distance routing under real-life operational conditions. Thus, being able to predict the actual stop sequence that a human driver would follow can help to improve route planning in last-mile delivery. This paper proposes a pair-wise attention-based pointer neural network for this prediction task using drivers' historical delivery trajectory data. In addition to the commonly used encoder-decoder architecture for sequence-to-sequence prediction, we propose a new attention mechanism based on an alternative specific neural network to capture the local pair-wise information for each pair of stops. To further capture the global efficiency of the route, we propose a new iterative sequence generation algorithm that is used after model training to identify the first stop of a route that yields the lowest operational cost. Results from an extensive case study on real operational data from Amazon's last-mile delivery operations in the US show that our proposed method can significantly outperform traditional optimization-based approaches and other machine learning methods (such as the Long Short-Term Memory encoder-decoder and the original pointer network) in finding stop sequences that are closer to high-quality routes executed by experienced drivers in the field. Compared to benchmark models, the proposed model can increase the average prediction accuracy of the first four stops from around 0.2 to 0.312, and reduce the disparity between the predicted route and the actual route by around 15%.
