A data driven hybrid heuristic for the dial-a-ride problem with time windows

The Dial-A-Ride Problem (DARP) consists of designing pick-up and delivery vehicle routing schedules for a set of customers. Commonly, it arises in door-to-door transportation for impaired or elderly people. The main objective is to accommodate as many users as possible with a minimum operation cost. It adds realistic precedence and transit time constraints on the pairing of vehicles and customers. This paper proposes a hybrid evolutionary heuristic for the dial-a-ride problem with time windows (DARPTW). It combines evolutionary crossover operators with a hybridized VNS algorithm. Data extracted from routes with DARP infeasibilities are used to guide customized local search operators. Using route duration minimization as its main objective, the competitiveness of the proposed heuristic is tested on benchmark instances.