Huang Jian, Zhang Yuxin, Xu Mengya, Lv Youyue, Zhang Junxian, Shafieezadeh M Mehdi
College of Physical Education, Jiangxi Normal University, Nanchang, 330022, China.
Department of Chemical Engineering, Islamic Azad University, Shiraz, Iran.
Sci Rep. 2025 Jul 31;15(1):27913. doi: 10.1038/s41598-025-13973-6.
Practical and objective player evaluation is a critical challenge in modern sports analytics, particularly in football, where selection decisions must balance multiple performance dimensions under uncertainty. This study introduces a hybrid Multi-Criteria Decision-Making (MCDM) methodology combining the Fuzzy Analytical Hierarchy Process (FAHP) and the Fuzzy Technique for Order Preference by Similarity to Ideal Solution (FTOPSIS) to evaluate and rank football players. The methodology integrates 12 criteria across three key domains: anthropometric attributes, fitness levels, and technical skills. FAHP is employed to derive the relative weights of criteria based on expert pairwise comparisons, while FTOPSIS calculates a closeness coefficient (CC) for each player to determine their ranking. Applied to a case study involving 20 youth football players aged 13-16, the results show that Player 10 achieved the highest closeness coefficient (CC = 0.92), followed by Player 4 (CC = 0.89) and Player 16 (CC = 0.85), indicating superior overall performance. Sensitivity analysis, including variations of ± 10% in criteria weights and performance scores, demonstrated the robustness of the top-ranked players' positions, with minimal fluctuation in CC values (e.g., Player 10's CC varied between 0.92 and 0.94). The proposed method provides a transparent and data-driven framework that accommodates variability in expert judgment and uncertainty in performance data. Despite strengths in robustness and comprehensiveness, limitations include dependence on expert input and scalability for larger datasets. Future work should explore the integration of position-specific evaluation models and their application to other sports domains. The findings highlight FAHP-FTOPSIS as a viable and rigorous approach for objective, multidimensional player selection in football.
在现代体育分析中，尤其是在足球领域，进行切实可行且客观的球员评估是一项关键挑战。在足球领域，选拔决策必须在不确定性下平衡多个表现维度。本研究引入了一种混合多准则决策（MCDM）方法，该方法结合了模糊层次分析法（FAHP）和模糊理想解逼近排序法（FTOPSIS）来评估足球运动员并对其进行排名。该方法整合了三个关键领域的12个标准：人体测量属性、体能水平和技术技能。FAHP用于基于专家成对比较得出标准的相对权重，而FTOPSIS则为每个球员计算一个贴近度系数（CC）以确定他们的排名。应用于一个涉及20名13 - 16岁青少年足球运动员的案例研究，结果表明10号球员获得了最高贴近度系数（CC = 0.92），其次是4号球员（CC = 0.89）和16号球员（CC = 0.85），表明其整体表现优异。敏感性分析，包括标准权重和表现得分±10%的变化，证明了排名靠前球员位置的稳健性，CC值波动极小（例如，10号球员的CC值在0.92和0.94之间变化）。所提出的方法提供了一个透明且数据驱动的框架，该框架能够适应专家判断的变异性和表现数据中的不确定性。尽管具有稳健性和全面性的优点，但局限性包括依赖专家输入以及对于更大数据集的可扩展性。未来的工作应探索特定位置评估模型的整合及其在其他体育领域的应用。研究结果突出了FAHP - FTOPSIS作为一种可行且严谨的方法，用于足球中客观、多维度的球员选拔。
