Scholarlink

The logistics performance of enterprises has been paramount in the recent years with improving transport. Consequently, the logistics performance of enterprises in their delivery of products has also assumed importance expected influence over all actors. The focus of this thesis is on the Vehicle Routing Problem (VRP) which is a realistic one for production management. The intent of the VRP is hierarchical - minimize vehicles and minimize travel costs. Along with the considerations of practical reality, metaheuristics will often be the most logical solution method to apply in applicable NP-hard problem settings such as the vehicle routing problem. In this thesis, a hybrid Genetic Algorithm (HGA) based on KNN machine learning and Genetic Algorithm is suggested to solve the vehicle routing problem. The HGA was assessed based on a series of reference instances set at varying levels. The computational results show that the developed HGA is not unreasonable in performance with regard to problem solving.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Hybrid Approach Based on K-Nearest Neighbors Machine Learning and Genetic Algorithm for a Logistic Chain Problem

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Hamza Ettakifi,
  • Said Tkatek

摘要

The logistics performance of enterprises has been paramount in the recent years with improving transport. Consequently, the logistics performance of enterprises in their delivery of products has also assumed importance expected influence over all actors. The focus of this thesis is on the Vehicle Routing Problem (VRP) which is a realistic one for production management. The intent of the VRP is hierarchical - minimize vehicles and minimize travel costs. Along with the considerations of practical reality, metaheuristics will often be the most logical solution method to apply in applicable NP-hard problem settings such as the vehicle routing problem. In this thesis, a hybrid Genetic Algorithm (HGA) based on KNN machine learning and Genetic Algorithm is suggested to solve the vehicle routing problem. The HGA was assessed based on a series of reference instances set at varying levels. The computational results show that the developed HGA is not unreasonable in performance with regard to problem solving.