Strategic insights from clustering analysis of essential oil sales in UMKM: A comprehensive study on product types, sizes, couriers, and distribution across Indonesian Provinces

Novianti Madhona Faizah (Universitas Tama Jagakarsa, Jakarta, Indonesia)
Desyi Erawati (Universitas Tama Jagakarsa, Indonesia)
Shirlyani Shirlyani (Universitas Tama Jagakarsa, Indonesia)
Luky Fabrianto (Universitas Nusa Mandiri, Indonesia)
Tiwuk Wahyuli Prihandayani (Universitas Mercu Buana, Indonesia)

Jurnal Mantik
Vol. 7 No. 4 (2024): February: Manajemen, Teknologi Informatika dan Komunikasi (Mantik)

This study explores the sales transactions of a Micro, Small, and Medium Enterprise (UMKM) that sells over 40 types of essential oils, totaling 2305 items sold in 2023. The products, packaged in small bottles (10-50 ml), were distributed to almost every province in Indonesia. The main objective is to cluster the data based on variables such as oil type, bottle size, courier company, and destination province. The elbow method determined an optimal number of clusters (k=4), and the Silhouette Coefficient validated the effectiveness of the clustering (0.7614). To simplify the complex clustering results, Principal Component Analysis (PCA) was used for visualization, providing a clear representation of 5 variables and 4 clusters. This study offers valuable insights for informed decision-making in UMKM's service enhancement and development