COAL TRADE DATA CLUSTERING USING K-MEANS (CASE STUDY PT. GLOBAL BANGKIT UTAMA)

Abstract

To compete in the business world, especially in the distributor fields, the company  must find strategy to increase the trade of products , one of them is through the analysis of trade data.  PT Global Bangkit Utama is a company engaged in coal distributor, which has many competitors. To face the competition, PT Global Bangkit Utama tries to find the right strategy. To make strategic decisions, The company analyzes the information on trades data. The data used in this study were coal trade data PT Global Bangkit Utama from January 2015 to August 2016.  One  method was Data Mining to determine the patterns of extracting information using the Clustering method.  The method clusters the objects which have similar characteristics to find the desired patterns.  The process of determining the patterns of clustering used  K-Means Algorithm.  K-Means algorithm is a clustering algorithm of the data with the partition system.  K-Means algorithm was chosen because it has a high level of accuracy and effectivity and require a relatively fast execution time due to its linerity.  This research produces 8 clusters using Elbow method. There is a characteristic equation in each cluster in the optimal cluster that will be used as business strategy determination. The business strategy obtained is to optimize distributors in the city of Karanganyar and make a storage place for coal..