A Classification-based Heuristic Approach for the Flying Sidekick Traveling Salesman Problem

This work proposes a novel heuristic approach for the Flying Sidekick Traveling Salesman Problem, that represents the first truck-and-drone routing problem defined in the literature. This approach integrates data science and machine learning techniques with combinatorial optimization methods. The aim is to determine a good/optimal customer-to-vehicle assignment a priori, reducing the solution space of the truck-and-drone routing problem. An extensive computational campaign on benchmark instances has been conducted to evaluate the effectiveness of the proposed approach.