A hybrid GRASP and VND heuristic for vehicle routing problem with dynamic requests

Abstract

This paper describes a hybrid heuristic that integrates the Greedy Randomized Adaptive Search Procedure (GRASP) and Variable Neighborhood Descent (VND) to address the Vehicle Routing Problem with Dynamic Requests (VRPDR). The VRPDR, a dynamic offshoot of the classical Vehicle Routing Problem (VRP), features customer requests emerging over time, with the objective of minimizing the total travel distance by devising a set of routes to serve all customers. The proposed method initially employs GRASP to construct an initial solution, followed by VND for exploration and refinement. The hybrid approach aims to utilize the strengths of both algorithms. Through testing on two sets of benchmark instances, namely dynamic pickup instances and dynamic delivery instances, 15 new optimal solutions are identified for the former and 11 for the latter. These results clearly demonstrate that the proposed algorithm competes favorably with the algorithms documented in the literature.