English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In last-mile delivery, drivers frequently deviate from planned deliveryroutes because of their tacit knowledge of the road and curbsideinfrastructure, customer availability, and other characteristics of therespective service areas. Hence, the actual stop sequences chosen by anexperienced human driver may be potentially preferable to the theoreticalshortest-distance routing under real-life operational conditions. Thus, beingable to predict the actual stop sequence that a human driver would follow canhelp to improve route planning in last-mile delivery. This paper proposes apair-wise attention-based pointer neural network for this prediction task usingdrivers' historical delivery trajectory data. In addition to the commonly usedencoder-decoder architecture for sequence-to-sequence prediction, we propose anew attention mechanism based on an alternative specific neural network tocapture the local pair-wise information for each pair of stops. To furthercapture the global efficiency of the route, we propose a new iterative sequencegeneration algorithm that is used after model training to identify the firststop of a route that yields the lowest operational cost. Results from anextensive case study on real operational data from Amazon's last-mile deliveryoperations in the US show that our proposed method can significantly outperformtraditional optimization-based approaches and other machine learning methods(such as the Long Short-Term Memory encoder-decoder and the original pointernetwork) in finding stop sequences that are closer to high-quality routesexecuted by experienced drivers in the field. Compared to benchmark models, theproposed model can increase the average prediction accuracy of the first fourstops from around 0.2 to 0.312, and reduce the disparity between the predictedroute and the actual route by around 15%.

Predicting Drivers' Route Trajectories in Last-Mile Delivery Using A Pair-wise Attention-based Pointer Neural Network