CV. Jaya Tunggal Keramik is a company that sale of ceramics. CV. Jaya Tunggal Keramik experienced some problems regarding ceramics and customers such as difficulties in sale ceramics to customers so that some ceramic products accumulate in the warehouse, such as being damaged and ceramic display becomes less good because it is stored too long and difficulty retaining customers because some customers do not want to order ceramic products. Lack of precise decision taken by the management CV. Jaya Tunggal Keramik in determining the strategy to supply ceramic and how to make it CV. Jaya Tunggal Keramik is difficult to estimate the stock of ceramic products to be provided and it is difficult to determine which potential customers can be maintained as a regular customer. This research uses K-Means algorithm. K-Means algorithm is a partitioning clustering method that separates data into different groups with iterative partitioning. By using this application, users can find out the estimated stock and price of ceramics as well as information about potential customers. Testing in this research using data of November 2017 that compared with data of December 2017. Based on ceramic data test results, there are some ceramics that are not in accordance with the predicted results so it can be concluded that the K-Means algorithm on the test inventory data inventory in this study is not fully can provide accurate estimates, this is because the use of the K-Means algorithm is strongly influenced by the cluster center results and the attributes used.
A Contraction-based Ratio-cut Partitioning Algorithm
