Scholarlink

<p>The m-polar fuzzy (mF) PROMETHEE methodology provides a robust framework for solving multi-criteria group decision-making (MCGDM) problems by combining the flexibility of fuzzy sets with the outranking strength of PROMETHEE. Despite its growing applications, in the literature, no attention has been given to the effect of data normalization on m-polar fuzzy PROMETHEE method performance. This study systematically examines five normalization techniques—linear sum, linear max, max–min, vector, and logarithmic—using both single-valued and fuzzy inputs. Two case studies, electric vehicle (EV) selection and tourist destination selection, are employed to evaluate the sensitivity of rankings to normalization. The novelty of this work is in the use of m-polar fuzzy PROMETHEE method to MCDM and MCGDM problems. EV selection problem provides the single value input for the m-polar fuzzy PROMETHEE approach while the tourist destination provides the fuzzy input for the algorithm. Results show that the choice of normalization significantly influences the final rank order, with linear sum normalization yielding the most stable rankings for EV selection, while nonlinear methods introduce rank reversals in both domains. Comparative analysis using Spearman’s rank correlation validates these findings. The research highlights the importance of carefully selecting normalization methods to ensure reliable decision outcomes and extends the applicability of the mF PROMETHEE framework to industrial and service sectors, including transportation, tourism, and beyond.</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

Assessing Data Normalization Effects on Electric Vehicle and Tourist Destination Selection: An m-Polar Fuzzy PROMETHEE Perspective

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月24日

查看全文
  • Madan Jagtap,
  • Prasad Karande

摘要

The m-polar fuzzy (mF) PROMETHEE methodology provides a robust framework for solving multi-criteria group decision-making (MCGDM) problems by combining the flexibility of fuzzy sets with the outranking strength of PROMETHEE. Despite its growing applications, in the literature, no attention has been given to the effect of data normalization on m-polar fuzzy PROMETHEE method performance. This study systematically examines five normalization techniques—linear sum, linear max, max–min, vector, and logarithmic—using both single-valued and fuzzy inputs. Two case studies, electric vehicle (EV) selection and tourist destination selection, are employed to evaluate the sensitivity of rankings to normalization. The novelty of this work is in the use of m-polar fuzzy PROMETHEE method to MCDM and MCGDM problems. EV selection problem provides the single value input for the m-polar fuzzy PROMETHEE approach while the tourist destination provides the fuzzy input for the algorithm. Results show that the choice of normalization significantly influences the final rank order, with linear sum normalization yielding the most stable rankings for EV selection, while nonlinear methods introduce rank reversals in both domains. Comparative analysis using Spearman’s rank correlation validates these findings. The research highlights the importance of carefully selecting normalization methods to ensure reliable decision outcomes and extends the applicability of the mF PROMETHEE framework to industrial and service sectors, including transportation, tourism, and beyond.