Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine Learning

Setting parameter values is crucial for the performance of metaheuristics. Tuning the parameters of a metaheuristic is a computationally costly task. Moreover, parameter tuning is difficult considering their inherent stochasticity and problem instance dependence. In this work, we explore the applica...

Full description

Saved in:
Bibliographic Details
Main Authors: Tomás Barros-Everett, Elizabeth Montero, Nicolás Rojas-Morales
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/6/2946
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850090084818223104
author Tomás Barros-Everett
Elizabeth Montero
Nicolás Rojas-Morales
author_facet Tomás Barros-Everett
Elizabeth Montero
Nicolás Rojas-Morales
author_sort Tomás Barros-Everett
collection DOAJ
description Setting parameter values is crucial for the performance of metaheuristics. Tuning the parameters of a metaheuristic is a computationally costly task. Moreover, parameter tuning is difficult considering their inherent stochasticity and problem instance dependence. In this work, we explore the application of machine learning algorithms to suggest suitable parameter values. We propose a methodology to use k-nearest neighbours and artificial neural network algorithms to predict suitable parameter values based on instance features. Here, we evaluate our proposal on the Capacitated Vehicle Routing Problem with Time Windows (CVRPTW) using its state-of-the-art algorithm, Hybrid Genetic Search (HGS). Additionally, we use the well-known tuning algorithm ParamILS to obtain suitable parameter configurations for HGS. We use a well-known instance set that considers between 200 and 1000 clients. Three sets of features based on geographical distribution, time windows, and client clustering are obtained. An in-depth exploratory analysis of the clustering features is also presented. The results are promising, demonstrating that the proposed method can successfully predict suitable parameter configurations for unseen instances and suggest configurations that perform better than baseline configurations. Furthermore, we present an explainability analysis to detect which features are more relevant for the prediction of suitable parameter values.
format Article
id doaj-art-e7f31ab7972a4baea0230afdfcbbca9c
institution DOAJ
issn 2076-3417
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-e7f31ab7972a4baea0230afdfcbbca9c2025-08-20T02:42:38ZengMDPI AGApplied Sciences2076-34172025-03-01156294610.3390/app15062946Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine LearningTomás Barros-Everett0Elizabeth Montero1Nicolás Rojas-Morales2Departamento de Informática, Universidad Técnica Federico Santa María, Avenida España 1680, Valparaíso 2390123, ChileDepartamento de Informática, Universidad Técnica Federico Santa María, Avenida España 1680, Valparaíso 2390123, ChileDepartamento de Informática, Universidad Técnica Federico Santa María, Avenida España 1680, Valparaíso 2390123, ChileSetting parameter values is crucial for the performance of metaheuristics. Tuning the parameters of a metaheuristic is a computationally costly task. Moreover, parameter tuning is difficult considering their inherent stochasticity and problem instance dependence. In this work, we explore the application of machine learning algorithms to suggest suitable parameter values. We propose a methodology to use k-nearest neighbours and artificial neural network algorithms to predict suitable parameter values based on instance features. Here, we evaluate our proposal on the Capacitated Vehicle Routing Problem with Time Windows (CVRPTW) using its state-of-the-art algorithm, Hybrid Genetic Search (HGS). Additionally, we use the well-known tuning algorithm ParamILS to obtain suitable parameter configurations for HGS. We use a well-known instance set that considers between 200 and 1000 clients. Three sets of features based on geographical distribution, time windows, and client clustering are obtained. An in-depth exploratory analysis of the clustering features is also presented. The results are promising, demonstrating that the proposed method can successfully predict suitable parameter configurations for unseen instances and suggest configurations that perform better than baseline configurations. Furthermore, we present an explainability analysis to detect which features are more relevant for the prediction of suitable parameter values.https://www.mdpi.com/2076-3417/15/6/2946automatic metaheuristic configurationexplainable artificial intelligencemachine learningparameter values prediction
spellingShingle Tomás Barros-Everett
Elizabeth Montero
Nicolás Rojas-Morales
Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine Learning
Applied Sciences
automatic metaheuristic configuration
explainable artificial intelligence
machine learning
parameter values prediction
title Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine Learning
title_full Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine Learning
title_fullStr Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine Learning
title_full_unstemmed Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine Learning
title_short Parameter Prediction for Metaheuristic Algorithms Solving Routing Problem Instances Using Machine Learning
title_sort parameter prediction for metaheuristic algorithms solving routing problem instances using machine learning
topic automatic metaheuristic configuration
explainable artificial intelligence
machine learning
parameter values prediction
url https://www.mdpi.com/2076-3417/15/6/2946
work_keys_str_mv AT tomasbarroseverett parameterpredictionformetaheuristicalgorithmssolvingroutingprobleminstancesusingmachinelearning
AT elizabethmontero parameterpredictionformetaheuristicalgorithmssolvingroutingprobleminstancesusingmachinelearning
AT nicolasrojasmorales parameterpredictionformetaheuristicalgorithmssolvingroutingprobleminstancesusingmachinelearning