Improvement through the use of the swap algorithm in a VRPTW case for 300 clients, generates efficient computational times

Abstract

Introduction: Vehicle routing scheduling with service fulfillment is a necessity for logistics companies seeking their competitive advantage. Objective: The objective of the following document is to determine the scheduling at the minimum cost of vehicle routing for a homogeneous fleet, complying with time windows. Method: Logistics costs, including the cost of operating vehicle routing, are a fundamental part of the decrease in costs, a direct increase in profits in companies, which is accompanied by creating a competitive advantage. The mathematical model used and the development of the combinatorial optimization algorithm for a VRPTW are exposed through the use of a savings matrix and swap heuristics.  Results: The savings algorithm establishes the initial construction phase based on the minimum distance, and the swap algorithm performs very fast exchanges through 12 interactions for a case of 300 clients, seeking to improve through the reduction of downtime, achieving savings of the total cost of 71.41% over the initial cost of the savings algorithm. Conclusions: For calculations of CVRPTW cases of 300 nodes, the swap algorithm achieves computational times of less than 30 seconds.