Halal Supply Chain Risk using Unsupervised Learning Methods for Clustering Leather Industries

Abstract

Cowhide plays a significant role in Indonesia's culinary and leather industries. It caters to the preferences of a predominantly Muslim population that emphasizes halal products. Regulatory authorities must understand its characteristics comprehensively to provide effective halal assurance to the diverse entities within Indonesia's leather industry. Traditional statistical methods for assessing halal compliance are inefficient due to the complexity and diversity of the leather industry's supply chain. This study addresses these challenges by employing unsupervised learning methods, specifically K-Means and Hierarchical clustering algorithms to analyze a dataset comprising 100 Cowhide Small and Medium Enterprises (SMEs) located in Garut Regency, West Java Province. This dataset includes 62 features that facilitate the clustering of these industries based on various halal risk factors. Experimental results indicate that the optimal number of clusters is four. The K-Means algorithm outperforms the Hierarchical clustering algorithm with a higher average silhouette score of 0.59 compared to 0.31. Furthermore, the K-Means algorithm demonstrates stability in clustering the data, making it a robust choice for this analysis. These clustering outcomes offer valuable insights into the SMEs operational characteristics and halal compliance risks, significantly enhancing the ability of regulatory authorities to implement effective halal assurance measures. Consequently, this study provides a robust framework for improving halal certification processes and aiding risk management within Indonesia's leather industry.