Presented by Ugur Arikan (Operations Research Scientist, Deutsche Post DHL) and Jonas Witt (Operations Research Scientist, Deutsche Post DHL)

Traditional last-mile delivery planning purely based on optimization methods often lacks important real-life aspects and thus does not satisfy relevant operational requirements. Only experienced couriers have tacit knowledge about the delivery area and its customers, forcing them to deviate from planned routes. They know where to find good parking spots, when to best approach certain business customers, and which neighbors to approach at what times when the actual recipient is not at home. This tacit knowledge is almost impossible to collect and maintain, let alone to incorporate in optimization algorithms.

Thus, we at Deutsche Post DHL developed an algorithm following a different approach: We aim at implicitly learning about this tacit knowledge from historical tours and combine this with optimization algorithms to plan routes that an experienced courier would choose. In this talk we will present details of our algorithm, which incorporates machine learning, statistics, and optimization in a novel way. Furthermore, we show how it impacted last-mile delivery planning at Deutsche Post after its rollout across Germany.