A New Method of Quality Evaluation for Electromagnet Products Based on SOM‐Kmeans and Enhanced TODIM

ABSTRACT In order to gain an advantage in the fierce market competition, enterprises need to establish a continuous and strict evaluation system for product quality. In addition, the complexity of mechatronic products requires that the evaluation method must be able to reflect the quality information comprehensively. However, current approaches for quality evaluation often grapple with subjectivity and do not fully mine the quality data. A novel methodology for product quality evaluation grounded in clustering and decision principles is introduced in this study. Initially, a self‐organizing map (SOM) and K‐means clustering are applied to the quality data to determine the number of quality categories. Subsequently, principal component analysis (PCA) is employed to derive the decision weights for the quality assessment procedure. This step effectively reduces the complexity associated with quality evaluation data and enhances the clarity of the findings. Finally, the TODIM (abbreviation for interactive and multicriterial decision‐making in Portuguese) method is utilized to scrupulously score each quality category. The efficacy and utility of the proposed method are corroborated through an empirical case study. This research underscores that the introduced approach offers a scalable and highly adaptable solution for quality assessment of other products.