<![CDATA[Knowledge management (KM) is widely recognized as a strategic resource for Small and Medium Enterprises (SMEs), yet its implementation in craft SMEs is often heterogeneous and fragmented. Most existing studies measure the average effect of KM on firm outcomes, which may obscure differences in KM maturity among individual SMEs and limit the design of differentiated interventions. This study aims to map craft SMEs based on their KM performance using an unsupervised learning approach. Primary data were collected from 70 bag and luggage craft SMEs in Sidoarjo Regency, East Java, through structured questionnaires using a six-point Likert scale. The study employed purposive sampling, with respondents selected based on specific criteria: the SMEs had to operate in the bag and luggage craft sector, be active during the data collection period, meet SME characteristics, and be represented by owners, managers, or SME leaders directly involved in daily business activities and decision-making. Therefore, the sample provides an analytical representation of active bag and luggage craft SMEs in the observed context, although it is not intended for statistical generalization to all craft SMEs in Sidoarjo. The questionnaire measured four KM dimensions: knowledge creation, knowledge acquisition, knowledge sharing, and knowledge application. The K-Means clustering algorithm was applied with z-score normalization, while the optimal number of clusters was determined using the Elbow Method and validated through the Silhouette Coefficient. The analysis identified two optimal clusters (k = 2) with a silhouette score of 0.441. Cluster 1 represents SMEs with consistently strong KM performance, while Cluster 0 represents SMEs with relatively weak KM performance. Knowledge acquisition and knowledge sharing emerged as the most differentiating dimensions. The novelty of this study lies in providing an empirical, data-driven map of KM maturity among craft SMEs as a basis for targeted KM development programs.]]>
Copyrights © 2026
