Scholarlink

<p>In recent years, the rise and rapid development of new energy vehicles (NEVs) has driven the global vehicle industry to transition from traditional fuel-powered models to electric vehicles. The selection of electric vehicles involves uncertainty and complexity and can be viewed as a complex decision problem. Therefore, based on the pairwise preference graph neural similarity (PPGNS) calculation method, this paper proposes a decision framework for addressing vehicle selection problems. First, we represent each expert’s pairwise preferences as graphs and use a graph similarity neural network to predict the similarity between these graphs, assigning expert weights based on the similarity levels. Next, we introduce the Pythagorean fuzzy weighted arithmetic mean (PFWAM) aggregation operator to combine expert weights and individual evaluation information to obtain the group decision matrix. Then, we determine attribute importance using the criteria importance though inter-criteria correlation (CRITIC) method in a Pythagorean fuzzy environment and rank the alternatives using the extended TOPSIS. Finally, sensitivity and comparative analyses confirm the effectiveness and practicality of the proposed method in vehicle selection problems.</p>

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

CRITIC-TOPSIS Method Based on Pairwise Preference Graph Neural Similarity and Its Application in Vehicle Selection

  • 期刊论文

  • 发表日期：

    2025年2月18日

查看全文
  • Jianping Fan,
  • Yanlong Han,
  • Meiqin Wu

摘要

In recent years, the rise and rapid development of new energy vehicles (NEVs) has driven the global vehicle industry to transition from traditional fuel-powered models to electric vehicles. The selection of electric vehicles involves uncertainty and complexity and can be viewed as a complex decision problem. Therefore, based on the pairwise preference graph neural similarity (PPGNS) calculation method, this paper proposes a decision framework for addressing vehicle selection problems. First, we represent each expert’s pairwise preferences as graphs and use a graph similarity neural network to predict the similarity between these graphs, assigning expert weights based on the similarity levels. Next, we introduce the Pythagorean fuzzy weighted arithmetic mean (PFWAM) aggregation operator to combine expert weights and individual evaluation information to obtain the group decision matrix. Then, we determine attribute importance using the criteria importance though inter-criteria correlation (CRITIC) method in a Pythagorean fuzzy environment and rank the alternatives using the extended TOPSIS. Finally, sensitivity and comparative analyses confirm the effectiveness and practicality of the proposed method in vehicle selection problems.