Implementation of Data Mining to Determine Drug Stock Inventory at Enok Pharmacy Using the K-Means Clustering Method

Abstract

Data mining processing has grown very rapidly in adapting every form of data analysis. Basically data mining deals with data analysis and the use of software techniques to look for patterns and regularities in hidden data sets. Stock management that is carried out inaccurately and carelessly will lead to high and uneconomical storage costs, because there may be vacancies or excess of certain products. This will certainly be very detrimental to all business actors such as the health center of the Enok Pharmacy. The k-means method is one of the data mining techniques used to help design an effective inventory strategy or stock order by utilizing sales transaction data that is already available in the company. This technique aims to classify medicinal products sold at Apotek Enok into several clusters of transaction data which are generally large in size using the k-means algorithm. This study aims to apply the k-means algorithm, the data taken as a case example is drug sales transaction data at the Enok Pharmacy. The results of the analysis of this study using 20 data. Clustering of drug data carried out with the K-Means algorithm obtained the results of the cluster after doing the 3rd iteration, namely there is a group of drugs that use slow in cluster 1 which has 6 members, groups of drugs that use fast are in cluster 2 which has 14 members. This cluster search uses a web base to find out which products are slow in use and which drugs are used fast.