Evaluation of face masks quality features using Kano model and unsupervised machine learning technique

Abstract

Purpose

The purpose of this study was to find the key face mask features using Kano model in combination with a hierarchical cluster analysis based on customer satisfaction (CS) and preference.

Design/methodology/approach

This study used 171 responses collected from a self-administrated online survey with convenience sampling where respondents were asked about 16 different features of face masks.

Findings

The study revealed that, among 6 Kano categories, 15 features were categorized as “one dimensional” and only the high price fell under the “reverse” category but all features were not equally weighted by customers. The result also showed viral protection and comfortability were the most desired features by customers regardless of its price and the “color matching” feature can act both as “one dimension” and as “attractive” feature.

Research limitations/implications

This study will help face mask producers to drive their resources towards those features which customers value more by showing how to prioritize features even if they fall under the same category.

Originality/value

This study used customer satisfaction and dissatisfaction index along with an unsupervised machine learning tool to improve features classification based on Kano model. The findings of this study can be used to formulate future research studies.

Citation

Sobuj, M., Alam, M.A. and Zannat, A. (2023), "Evaluation of face masks quality features using Kano model and unsupervised machine learning technique", Research Journal of Textile and Apparel, Vol. 27 No. 4, pp. 560-572. https://doi.org/10.1108/RJTA-11-2021-0141