Reza Aditya, Bernieka Desyaranti, Fandi Achmad

Organizational Performance-Based Segmentation of Batik SMEs Using K-Means Clustering

Introduction

Organizational performance-based segmentation of batik smes using k-means clustering. Segment Batik SMEs in Indonesia based on organizational performance using K-Means clustering. Discover two distinct segments for targeted strategies & policy formulation.

Abstract

Batik Small and Medium Enterprises (SMEs) play a strategic role in the economy and cultural preservation in Indonesia, yet exhibit varying levels of organizational performance. This heterogeneity demands a more objective and data-driven segmentation approach so that development strategies and policies can be implemented effectively. This study aims to segment Batik SMEs based on organizational performance using the K-Means clustering method. Primary data were obtained through a structured questionnaire survey of 56 Batik SMEs, with organizational performance indicators covering operational and financial performance. All data were normalized using StandardScaler to ensure scale equality between variables. The optimal number of clusters was determined using the Elbow method and the silhouette coefficient. The analysis results showed that a two-cluster configuration was the optimal solution with the highest silhouette coefficient value. The K-Means model resulted in two segments of Batik SMEs with significantly different organizational performance characteristics: SMEs with low organizational performance and SMEs with high organizational performance. Centroid value analysis and cluster visualization confirmed clear cluster separation and a good level of internal homogeneity. These findings indicate that segmenting Batik SMEs based on organizational performance using an unsupervised learning approach is effective in uncovering the structure of performance heterogeneity. This research contributes to providing a data-driven segmentation framework that can be utilized as a tool to support managerial decision-making and policy formulation to sustainably improve the performance and competitiveness of Batik SMEs.

Review

The paper "Organizational Performance-Based Segmentation of Batik SMEs Using K-Means Clustering" tackles a highly relevant and practical issue concerning the heterogeneous performance of Batik Small and Medium Enterprises (SMEs) in Indonesia. By employing a data-driven approach, specifically K-Means clustering, the study aims to provide an objective segmentation of these critical economic and cultural entities. This initiative is commendable, as it directly responds to the need for more targeted development strategies and policies, moving beyond anecdotal observations to a more empirical understanding of performance variations. The study's objective is clearly articulated, and its potential to inform evidence-based interventions is significant. Methodologically, the study demonstrates a robust application of unsupervised machine learning techniques. The use of K-Means clustering, supported by appropriate pre-processing steps like StandardScaler normalization and optimal cluster determination using the Elbow method and silhouette coefficient, reflects a careful and systematic approach. The identification of two distinct segments – SMEs with low and high organizational performance – is a significant finding, providing actionable insights into the underlying structure of performance heterogeneity. The reliance on primary data collected through a structured questionnaire, covering both operational and financial indicators, adds credibility to the findings, even with a sample size of 56. The clear separation and internal homogeneity of the identified clusters, confirmed by centroid analysis and visualization, further strengthen the study's conclusions regarding the effectiveness of this segmentation framework. While the study presents valuable findings, some areas could be considered for future research or more detailed discussion. Expanding the sample size beyond 56 in future work would undoubtedly enhance the generalizability of the findings across the diverse landscape of Indonesian Batik SMEs. Additionally, a deeper qualitative or quantitative exploration into the *specific* contextual factors and business practices that characterize each performance segment, beyond general operational and financial indicators, could yield even richer insights for highly targeted interventions. Nevertheless, the research makes a substantial contribution by providing a concrete, data-driven framework for segmenting SMEs based on performance. This framework is highly valuable for supporting managerial decision-making, allowing for the customization of support programs, training, and policy formulation, ultimately fostering the sustainable improvement of competitiveness within the Batik SME sector.

Full Text

Comments

You need to be logged in to post a comment.