Predicting last-mile delivery route deviations using machine learning

• Regression and classification models quantify and predict route deviations. • Route deviation metrics to measure the similarity of routes. • Models use sequential route structure, route attributes, and driver information. • Exploring tacit driver knowledge for efficient route planning. • Publicly available dataset supports further research. Route planning in last-mile delivery is a complex task with many challenges, directly impacting delivery efficiency and costs. Drivers often deviate from optimized planned routes based on their knowledge. Using the properties of machine learning, this study aims to determine whether machine learning techniques can effectively predict deviations by drivers from planned routes and quantify the extent of such deviations. We propose to predict route deviations by analyzing a logistics company’s historical data of planned and actual routes using deep neural networks, with the dataset made publicly available. Our methodology incorporates both regression and classification models. The regression model estimates the degree of deviation, while the classification model aims to predict whether the deviation from a planned route will exceed a given threshold, based on different deviation metrics. As the input, we leverage the sequential structure of the route with route properties and drivers information. The computational experiments explore extending the given input to the models and testing various state-of-art neural network architectures. Our results demonstrate strong performance on both tasks, with our models achieving 9 − 19 % improvements in regression metrics and 3 − 15 % improvements in classification metrics compared to specified benchmarks, with statistical tests confirming the significance of these improvements.