PERBANDINGAN CLUSTERING ALGORITMA K-MEANS
DAN ALGORITMA K-MEDOIDS UNTUK MENGOPTIMALISASIKAN STOK BARANG ( STUDI KASUS : PT. JAYADI KIJANG KENCANA )
Total View This Week0
Total View This Week0
Institusion
Universitas Stikubank
Author
Mulyaningrum, Endah
Subject
QA76 Computer software
Datestamp
2022-05-20 02:58:10
Abstract :
Penimbunan stok barang berlebih di gudang menjadi awal permasalahan. Jumlah permintaan konsumen yang fluktuaktif mengakibatkan ketersediaan stok barang menjadi tidak stabil. Maka, perlu mengoptimalkan stok barang dengan mengelompokkan barang apa saja yang diminati menggunakan teknik data mining dengan metode clustering serta algoritma k-means dan algoritma k-medoids. Penelitian ini bertujuan mengimplementasikan serta membandingkan hasil klaster stok barang menggunakan algoritma k-means dan algoritma k-medoids. Perhitungan manual dan Rstudio menghasilkan 3 klaster. Implementasi kedua algoritma pada data stok barang berhasil dengan hasil rerata barang keluar algoritma k-means klaster 1 tertinggi bulan Mei sebesar 8,53, klaster 2 rerata barang keluar tertinggi bulan Januari sebesar 12,2 dan klaster 3 rerata barang keluar tertinggi bulan Mei sebesar 12,2. Sedangkan rerata barang keluar algoritma k-medoids klaster 1 tertinggi bulan April sebesar 11,9, klaster 2 rerata barang keluar tertinggi bulan Februari sebesar 10,2 dan klaster 3 rerata barang keluar tertinggi bulan Maret sebesar 8,83. The accumulation of excess stock in the warehouse is the beginning of the problem. The amount of fluctuating consumer demand resulted in unstable stock availability. So, it is necessary to optimize the stock of goods by grouping any items of interest using data mining techniques with the clustering method as well as the kmeans algorithm and the k-medoids algorithm. This study aims to implement and compare the results of inventory clusters using the k-means algorithm and the kmedoids algorithm. Manual calculations and Rstudio generate 3 clusters. The implementation of the two algorithms on stock data was successful with the result that the highest average of goods out of the k-means algorithm for cluster 1 was 8.53 in May, cluster 2 had the highest average outflow of goods in January of 12.2 and cluster 3 the highest average of goods out in May was 12. ,2. Meanwhile, the highest average outflow of k-medoids in April was 11.9, cluster 2 had the highest average outflow in February at 10.2, and cluster 3, the highest average outflow in March, was 8.83.
Penimbunan stok barang berlebih di gudang menjadi awal permasalahan. Jumlah permintaan konsumen yang fluktuaktif mengakibatkan ketersediaan stok barang menjadi tidak stabil. Maka, perlu mengoptimalkan stok barang dengan mengelompokkan barang apa saja yang diminati menggunakan teknik data mining dengan metode clustering serta algoritma k-means dan algoritma k-medoids. Penelitian ini bertujuan mengimplementasikan serta membandingkan hasil klaster stok barang menggunakan algoritma k-means dan algoritma k-medoids. Perhitungan manual dan Rstudio menghasilkan 3 klaster. Implementasi kedua algoritma pada data stok barang berhasil dengan hasil rerata barang keluar algoritma k-means klaster 1 tertinggi bulan Mei sebesar 8,53, klaster 2 rerata barang keluar tertinggi bulan Januari sebesar 12,2 dan klaster 3 rerata barang keluar tertinggi bulan Mei sebesar 12,2. Sedangkan rerata barang keluar algoritma k-medoids klaster 1 tertinggi bulan April sebesar 11,9, klaster 2 rerata barang keluar tertinggi bulan Februari sebesar 10,2 dan klaster 3 rerata barang keluar tertinggi bulan Maret sebesar 8,83. The accumulation of excess stock in the warehouse is the beginning of the problem. The amount of fluctuating consumer demand resulted in unstable stock availability. So, it is necessary to optimize the stock of goods by grouping any items of interest using data mining techniques with the clustering method as well as the kmeans algorithm and the k-medoids algorithm. This study aims to implement and compare the results of inventory clusters using the k-means algorithm and the kmedoids algorithm. Manual calculations and Rstudio generate 3 clusters. The implementation of the two algorithms on stock data was successful with the result that the highest average of goods out of the k-means algorithm for cluster 1 was 8.53 in May, cluster 2 had the highest average outflow of goods in January of 12.2 and cluster 3 the highest average of goods out in May was 12. ,2. Meanwhile, the highest average outflow of k-medoids in April was 11.9, cluster 2 had the highest average outflow in February at 10.2, and cluster 3, the highest average outflow in March, was 8.83.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
