基于信息增益的项目排序和q-学习的分子模糊节能建筑投资排序

IF 7.1 2区 环境科学与生态学 Q1 BIOTECHNOLOGY & APPLIED MICROBIOLOGY
Gang Kou , Hasan Dinçer , Yaşar Gökalp , Serhat Yüksel , Serkan Eti , Ümit Hacıoğlu
{"title":"基于信息增益的项目排序和q-学习的分子模糊节能建筑投资排序","authors":"Gang Kou ,&nbsp;Hasan Dinçer ,&nbsp;Yaşar Gökalp ,&nbsp;Serhat Yüksel ,&nbsp;Serkan Eti ,&nbsp;Ümit Hacıoğlu","doi":"10.1016/j.eti.2025.104378","DOIUrl":null,"url":null,"abstract":"<div><div>Energy-positive buildings have gained significant attention as a sustainable solution to the growing global energy crisis. However, the efficient allocation of limited resources remains major challenges for optimizing these investments. There is a need for a new priority analysis for these factors. This article aims to determine the appropriate investment strategies regarding energy positive building projects via a novel model. The relevant project factors are detected using the information gain-based attribute selection in the first stage. The balanced evaluation matrices are created by q-learning in the following section. The weights of performance indicators are computed by molecular fuzzy cognitive maps. Moreover, project alternatives for energy positive building investments are examined via molecular fuzzy multi-objective particle swarm optimization. The main contribution of this study to the literature is that prior investment strategies for the improvements of the energy positive building projects can be identified with the help of a novel decision-making model. The main superiority of the proposed methodology is calculation of the importance weights of the experts. The results illustrate that the information gain method reduces the initial eight project alternatives to five, with the highest information gain values (0.750 for energy production potential and 1.000 for the use of high-performance materials) highlighting the most influential factors. The q-learning algorithm balances expert evaluations, achieving convergence with a tolerance of 0.02, ensuring stability in decision matrices. The MF cognitive maps assign weights to criteria, with the use of high-performance materials (weight: 0.256) and technological infrastructure (weight: 0.253) emerging as the most critical. The MF-MOPSO ranking results show consistent performance across five molecular geometry shapes, with the vertical urban farming tower (average score: 0.1562) and net-positive educational campus (average score: 0.1560) as the top alternatives. The model’s superiority is further validated through comparative analysis with the ARAS method, confirming robustness against weight variations (1–2 % changes). These results provide actionable insights for policymakers and investors to allocate resources effectively, emphasizing high-performance materials and technological advancements as key drivers for energy-positive building investments.</div></div>","PeriodicalId":11725,"journal":{"name":"Environmental Technology & Innovation","volume":"40 ","pages":"Article 104378"},"PeriodicalIF":7.1000,"publicationDate":"2025-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Information-gain based project prioritization and q-learning molecular fuzzy ranking for energy positive building investments\",\"authors\":\"Gang Kou ,&nbsp;Hasan Dinçer ,&nbsp;Yaşar Gökalp ,&nbsp;Serhat Yüksel ,&nbsp;Serkan Eti ,&nbsp;Ümit Hacıoğlu\",\"doi\":\"10.1016/j.eti.2025.104378\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Energy-positive buildings have gained significant attention as a sustainable solution to the growing global energy crisis. However, the efficient allocation of limited resources remains major challenges for optimizing these investments. There is a need for a new priority analysis for these factors. This article aims to determine the appropriate investment strategies regarding energy positive building projects via a novel model. The relevant project factors are detected using the information gain-based attribute selection in the first stage. The balanced evaluation matrices are created by q-learning in the following section. The weights of performance indicators are computed by molecular fuzzy cognitive maps. Moreover, project alternatives for energy positive building investments are examined via molecular fuzzy multi-objective particle swarm optimization. The main contribution of this study to the literature is that prior investment strategies for the improvements of the energy positive building projects can be identified with the help of a novel decision-making model. The main superiority of the proposed methodology is calculation of the importance weights of the experts. The results illustrate that the information gain method reduces the initial eight project alternatives to five, with the highest information gain values (0.750 for energy production potential and 1.000 for the use of high-performance materials) highlighting the most influential factors. The q-learning algorithm balances expert evaluations, achieving convergence with a tolerance of 0.02, ensuring stability in decision matrices. The MF cognitive maps assign weights to criteria, with the use of high-performance materials (weight: 0.256) and technological infrastructure (weight: 0.253) emerging as the most critical. The MF-MOPSO ranking results show consistent performance across five molecular geometry shapes, with the vertical urban farming tower (average score: 0.1562) and net-positive educational campus (average score: 0.1560) as the top alternatives. The model’s superiority is further validated through comparative analysis with the ARAS method, confirming robustness against weight variations (1–2 % changes). These results provide actionable insights for policymakers and investors to allocate resources effectively, emphasizing high-performance materials and technological advancements as key drivers for energy-positive building investments.</div></div>\",\"PeriodicalId\":11725,\"journal\":{\"name\":\"Environmental Technology & Innovation\",\"volume\":\"40 \",\"pages\":\"Article 104378\"},\"PeriodicalIF\":7.1000,\"publicationDate\":\"2025-07-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Environmental Technology & Innovation\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2352186425003645\",\"RegionNum\":2,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"BIOTECHNOLOGY & APPLIED MICROBIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Environmental Technology & Innovation","FirstCategoryId":"93","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2352186425003645","RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BIOTECHNOLOGY & APPLIED MICROBIOLOGY","Score":null,"Total":0}
引用次数: 0

摘要

节能建筑作为应对日益严重的全球能源危机的可持续解决方案,受到了广泛关注。然而,有限资源的有效分配仍然是优化这些投资的主要挑战。有必要对这些因素进行新的优先分析。本文旨在通过一种新的模型来确定节能建筑项目的适当投资策略。在第一阶段,使用基于信息增益的属性选择来检测相关的项目因素。平衡评估矩阵将在下一节中通过q学习创建。利用分子模糊认知图计算性能指标的权重。此外,采用分子模糊多目标粒子群优化方法对节能建筑投资方案进行了优选。本研究对文献的主要贡献在于,利用一种新的决策模型,可以识别出节能建筑项目改进的先验投资策略。该方法的主要优点是计算专家的重要权重。结果表明,信息增益法将最初的8个项目备选方案减少到5个,其中最高的信息增益值(能源生产潜力为0.750,高性能材料使用为1.000)突出了影响因素。q-learning算法平衡了专家评估,实现了0.02的收敛容差,确保了决策矩阵的稳定性。MF认知地图为标准分配权重,其中高性能材料(权重:0.256)和技术基础设施(权重:0.253)的使用是最关键的。MF-MOPSO的排名结果显示,五种分子几何形状的表现一致,其中垂直城市农业塔(平均得分:0.1562)和净正性教育校园(平均得分:0.1560)是最佳选择。通过与ARAS方法的对比分析,进一步验证了模型的优越性,证实了对权重变化(1-2 %变化)的稳健性。这些结果为决策者和投资者有效分配资源提供了可操作的见解,强调高性能材料和技术进步是节能建筑投资的关键驱动因素。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Information-gain based project prioritization and q-learning molecular fuzzy ranking for energy positive building investments
Energy-positive buildings have gained significant attention as a sustainable solution to the growing global energy crisis. However, the efficient allocation of limited resources remains major challenges for optimizing these investments. There is a need for a new priority analysis for these factors. This article aims to determine the appropriate investment strategies regarding energy positive building projects via a novel model. The relevant project factors are detected using the information gain-based attribute selection in the first stage. The balanced evaluation matrices are created by q-learning in the following section. The weights of performance indicators are computed by molecular fuzzy cognitive maps. Moreover, project alternatives for energy positive building investments are examined via molecular fuzzy multi-objective particle swarm optimization. The main contribution of this study to the literature is that prior investment strategies for the improvements of the energy positive building projects can be identified with the help of a novel decision-making model. The main superiority of the proposed methodology is calculation of the importance weights of the experts. The results illustrate that the information gain method reduces the initial eight project alternatives to five, with the highest information gain values (0.750 for energy production potential and 1.000 for the use of high-performance materials) highlighting the most influential factors. The q-learning algorithm balances expert evaluations, achieving convergence with a tolerance of 0.02, ensuring stability in decision matrices. The MF cognitive maps assign weights to criteria, with the use of high-performance materials (weight: 0.256) and technological infrastructure (weight: 0.253) emerging as the most critical. The MF-MOPSO ranking results show consistent performance across five molecular geometry shapes, with the vertical urban farming tower (average score: 0.1562) and net-positive educational campus (average score: 0.1560) as the top alternatives. The model’s superiority is further validated through comparative analysis with the ARAS method, confirming robustness against weight variations (1–2 % changes). These results provide actionable insights for policymakers and investors to allocate resources effectively, emphasizing high-performance materials and technological advancements as key drivers for energy-positive building investments.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Environmental Technology & Innovation
Environmental Technology & Innovation Environmental Science-General Environmental Science
CiteScore
14.00
自引率
4.20%
发文量
435
审稿时长
74 days
期刊介绍: Environmental Technology & Innovation adopts a challenge-oriented approach to solutions by integrating natural sciences to promote a sustainable future. The journal aims to foster the creation and development of innovative products, technologies, and ideas that enhance the environment, with impacts across soil, air, water, and food in rural and urban areas. As a platform for disseminating scientific evidence for environmental protection and sustainable development, the journal emphasizes fundamental science, methodologies, tools, techniques, and policy considerations. It emphasizes the importance of science and technology in environmental benefits, including smarter, cleaner technologies for environmental protection, more efficient resource processing methods, and the evidence supporting their effectiveness.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信