Evaluating Effectiveness of Crossover Operator in Genetic Algorithm for Travelling Salesman Problem using Regression Machine Learning: Post-Experiment Analysis

Singgih, Ivan Kristianto and Prabowo, Aditya Rio and Meilanitasari, Prita (2026) Evaluating Effectiveness of Crossover Operator in Genetic Algorithm for Travelling Salesman Problem using Regression Machine Learning: Post-Experiment Analysis. Industrial Engineering & Management Systems, 25 (3). pp. 571-582. ISSN 1598-7248

PDF Documents_TSP_ML.pdf Download (12MB)

Abstract

Traveling salesman problem (TSP) is an important problem because it is the base of many real-world transportation problems. One of the most effective and frequently used optimization methods for solving TSP is the genetic algorithm. Generally, genetic algorithms employ two types of operators: crossover and mutation. Most researchers apply the crossover operator to genetic algorithms to find better solutions, without analyzing the improvement in solution quality each time the crossover is applied. Such an approach is commonly used with the aim of simply finding the best route, for example, by comparing the quality of final solutions produced by the genetic algorithm and other methods. Unlike previous studies, this study examines the characteristics of the updated solution within the genetic algorithm run, which is modified by crossover, specifically, the improvement in solution quality. To the best knowledge of the authors, this study is the first to evaluate the effectiveness of crossover operators using regression machine learning methods. The numerical experiments show that the most important feature to predict the change in the best objective values is the distance between customers outside the cutting points, followed by the distance between customers with in the cutting points, then the number of customers (outside then between the cutting points). This study provides in sights for researchers to develop more effective solution methods. Keywords: Travelling Salesman Problem, Genetic Algorithm, Regression Machine Learning, Crossover

Item Type:	Article
Subjects:	Q Science > Q Science (General)
Divisions:	Faculty of Engineering > Department of Industrial Engineering
Depositing User:	IVAN KRISTIANTO SINGGIH
Date Deposited:	21 Oct 2026 01:40
Last Modified:	26 Oct 2026 02:32
URI:	http://repository.ubaya.ac.id/id/eprint/51388

Actions (login required)

View Item