汽车工业
工程类
汽车工程
计算机科学
制造工程
工程制图
机械工程
系统工程
法律工程学
产品(数学)
标识
DOI:10.1080/00405000.2026.2736894
摘要
Amid the green transformation of the automotive industry, sustainable seat fabrics require effective CMF evaluation methods linking design features with user perception. This study proposes a multimodal evaluation framework integrating Kansei engineering, AI-based emotion perception, and the fuzzy analytic hierarchy process (FAHP) to quantify the relationship between rPET automotive seat fabric characteristics and user responses. The framework establishes CMF feature weights through FAHP, combines subjective Kansei evaluation with objective facial expression analysis, and integrates both data sources through a weighted evaluation model. Results show that hand-feel treatment (0.121), weaving structure (0.116), and color matching (0.115) are the most influential features. Sample A02 achieved the highest overall evaluation, and subjective and objective rankings showed a positive correlation (Spearman’s ρ = 0.70). The proposed framework provides quantitative support for sustainable CMF design optimization.
科研通智能强力驱动
Strongly Powered by AbleSci AI