Implementasi Data Mining Untuk Pengelompokan Aset Perusahaan Pada Regional Office PT.NAK Dengan Metode K-Means Clustering

Jumadi; Hidayatullah Al Islami

Authors

Jumadi Universitas Pamulang
Hidayatullah Al Islami Universitas Pamulang

Keywords:

Assets, Data Mining, K-Means, Clustering, Regional Office

Abstract

In terms of the procurement and distribution of company assets in the PT.NAK regional office, it is influenced by the planning carried out. So that assets can be available with the type and amount that is sufficient according to the operational needs of the project. At present the location of the existing regional office area has not carried out activities as expected, considering several obstacles including proper data collection and monitoring of assets so that the whereabouts of assets are not known and properly managed, the ability of experts, limited knowledge of regional office management, policies head office management, and limited access to facilities. As a result of this condition, asset management is still conventional where data processed manually is considered ineffective in terms of time and does not provide clear information regarding asset data that must be prioritized for replacement or repair. This research will develop a data mining system that can help the process of grouping regional office asset data into several clusters, thereby reducing the time it takes to process data for plans to purchase assets as needed and reduce the accumulation of assets that are not used quickly. This study uses the K-Means clustering method, which is a method that can be used to group data based on the degree of similarity and characteristics of the data.

The application of the K-Means method in this study is used to process data that will be entered into each of the 3 clusters, the first cluster is asset data that is not suitable for use or must be repaired or replaced with purchases over 5 years, and the second is that asset data is still feasible used with a maximum purchase of 5 years and the three data assets are very suitable for use with a maximum of 2 years of new purchases. The data used is data obtained from research results in the field where there are 6434 raw data that are processed according to research needs into 322 data that will be tested on the system. Based on the test results on the system using the K-Means Clustering method, it was found that 119 assets were included in the first cluster, 5 assets were included in the second cluster and 199 assets were included in the third cluster.

References

Jollyta, D., William, R., & Muhammad, Z., 2020, Konsep Data Mining Dan Penerapannya, Deepublish, Yogyakarta.

Siregar, A.M., & Adam, P., 2020, Pengolahan Data Menjadi Informasi Denga RapidMiner, Kekata Group CV, Indonesia.

Najia, S. 2019, Klasifikasi Barang Menggunakan Metode Clustering K-Means Dalam Penentuan Prediksi Stok Barang UKM Mar’ah Jilbab Kediri. Skripsi. Universitas Islam Negeri Maulana Malikibrahim, Malang.

Rega, F.P.D. 2019, Penerapan Data Mining Menggunakan Algoritma K-Means Untuk Klasifikasi Penyakit ISPA. Skripsi. Universitas Pelita Bangsa, Bekasi.

Muhammad, L.H. 2020, Pengelompokan Data Transaksi Penjualan Menggunakan Algoritma Clustering K-Means Pada Toko AN NUR. Skripsi. Universitas Semarang, Semarang.

Efraim, H.B. 2020, Pengelompokan Data Obat Menggunakan Metode K-Means Clustering Pada UPT Puskesmas Kondoran Kec.Sangala. Skripsi. Institut Teknologi Nasional, Malang.

Event. 2021, Analisi Clustering Dengan K-Means Untuk Pengelompokan Penjualan Produk Pada Hotel Newton. Skripsi. Universitas Putera Batam, Batam.

Uun, H. 2022, Implementasi Metode K-Means Clustering Untuk Klasifikasi Penjualan Parfum. Skripsi. Universitas Nusantara, Kediri.