Taguchi-Grey Relational Analysis Method for Parameter Tuning of Multi-objective Pareto Ant Colony System Algorithm

Q4 Computer Science

International Journal of Information and Communication Technology Pub Date : 2023-04-03 DOI:10.32890/jict2023.22.2.1

Shatha Abdulhadi Muthana, K. Ku-Mahamud

{"title":"Taguchi-Grey Relational Analysis Method for Parameter Tuning of Multi-objective Pareto Ant Colony System Algorithm","authors":"Shatha Abdulhadi Muthana, K. Ku-Mahamud","doi":"10.32890/jict2023.22.2.1","DOIUrl":null,"url":null,"abstract":"In any metaheuristic, the parameter values strongly affect the efficiency of an algorithm’s search. This research aims to find the optimal parameter values for the Pareto Ant Colony System (PACS) algorithm, which is used to obtain solutions for the generator maintenance scheduling problem. For optimal maintenance scheduling with low cost, high reliability, and low violation, the parameter values of the PACS algorithm were tuned using the Taguchi and Gray Relational Analysis (Taguchi-GRA) method through search-based approach. The new parameter values were tested on two systems. i.e., 26- and 36-unit systems for window with operational hours [3000-5000]. The gray relational grade (GRG) performance metric and the Friedman test were used to evaluate the algorithm’s performance. The Taguchi-GRA method that produced the new values for the algorithm’s parameters was shown to be able to provide a better multi-objective generator maintenance scheduling (GMS) solution. These values can be benchmarked in solving multi-objective GMS problems using the multi-objective PACS algorithm and its variants.","PeriodicalId":39396,"journal":{"name":"International Journal of Information and Communication Technology","volume":"13 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Information and Communication Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.32890/jict2023.22.2.1","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Computer Science","Score":null,"Total":0}

引用次数: 0

Abstract

In any metaheuristic, the parameter values strongly affect the efficiency of an algorithm’s search. This research aims to find the optimal parameter values for the Pareto Ant Colony System (PACS) algorithm, which is used to obtain solutions for the generator maintenance scheduling problem. For optimal maintenance scheduling with low cost, high reliability, and low violation, the parameter values of the PACS algorithm were tuned using the Taguchi and Gray Relational Analysis (Taguchi-GRA) method through search-based approach. The new parameter values were tested on two systems. i.e., 26- and 36-unit systems for window with operational hours [3000-5000]. The gray relational grade (GRG) performance metric and the Friedman test were used to evaluate the algorithm’s performance. The Taguchi-GRA method that produced the new values for the algorithm’s parameters was shown to be able to provide a better multi-objective generator maintenance scheduling (GMS) solution. These values can be benchmarked in solving multi-objective GMS problems using the multi-objective PACS algorithm and its variants.

查看原文本刊更多论文

多目标Pareto蚁群算法参数整定的田口灰关联分析方法

在任何元启发式算法中，参数值都强烈影响算法的搜索效率。本研究旨在寻找帕累托蚁群系统(Pareto Ant Colony System, PACS)算法的最优参数值，并将其用于求解发电机维修调度问题。为了实现低成本、高可靠性、低违章的最优维修调度，采用基于搜索的田口灰色关联分析(Taguchi- gra)方法对PACS算法的参数值进行了调整。在两个系统上对新参数值进行了测试。即，对于运行时间[3000-5000]的窗口，采用26单元和36单元系统。采用灰色关联度(GRG)性能指标和Friedman检验来评价算法的性能。结果表明，采用Taguchi-GRA方法对算法参数产生新值，能够提供较好的多目标发电机维修调度方案。这些值可以在使用多目标PACS算法及其变体解决多目标GMS问题时作为基准。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

International Journal of Information and Communication Technology Computer Science-Information Systems

CiteScore

0.70

自引率

0.00%

发文量

期刊介绍： IJICT is a refereed journal in the field of information and communication technology (ICT), providing an international forum for professionals, engineers and researchers. IJICT reports the new paradigms in this emerging field of technology and envisions the future developments in the frontier areas. The journal addresses issues for the vertical and horizontal applications in this area. Topics covered include: -Information theory/coding- Information/IT/network security, standards, applications- Internet/web based systems/products- Data mining/warehousing- Network planning, design, administration- Sensor/ad hoc networks- Human-computer intelligent interaction, AI- Computational linguistics, digital speech- Distributed/cooperative media- Interactive communication media/content- Social interaction, mobile communications- Signal representation/processing, image processing- Virtual reality, cyber law, e-governance- Microprocessor interfacing, hardware design- Control of industrial processes, ERP/CRM/SCM