Using Recursive KMeans and Dijkstra Algorithm to Solve CVRP

TL;DR

This paper proposes a novel recursive K-Means clustering combined with Dijkstra's algorithm to approximate solutions for the NP-Hard Capacitated Vehicle Routing Problem, aiming to improve routing efficiency.

Contribution

It introduces an integrated approach using recursive K-Means and Dijkstra's algorithm to address CVRP, which is a new combination for this problem.

Findings

Demonstrates the feasibility of combining K-Means and Dijkstra for CVRP

Provides a framework for near-optimal routing solutions

Lays groundwork for future implementation and testing

Abstract

Capacitated vehicle routing problem (CVRP) is being one of the most common optimization problems in our days, considering the wide usage of routing algorithms in multiple fields such as transportation domain, food delivery, network routing, ... Capacitated vehicle routing problem is classified as an NP-Hard problem, hence normal optimization algorithm can't solve it. In our paper, we discuss a new way to solve the mentioned problem, using a recursive approach of the most known clustering algorithm "K-Means", one of the known shortest path algorithm "Dijkstra", and some mathematical operations. In this paper, we will show how to implement those methods together in order to get the nearest solution of the optimal route, since research and development are still on go, this research paper may be extended with another one, that will involve the implementational results of this thoric side.