Implementation of Data Mining Grouping Product Order Distribution Center In PT. Indomarco Prismatama Medan Branch

- Data mining is the technique of extracting previously unknown information in a set of data in the database. Data mining has been applied in various fields which require extracting information. One of them in groupings of data. Grouping is used to divide a set of data into several sections that are useful to more easily identify a class of data. Distribution companies can use groupings of one to determine the intensity of the volume of goods ordered. The study analyzes the application of data mining algorithms k-meansclustering to elicit information from the data ordering goods contained in centerPT distribution. Indomarco Prismatama Medan branch. That is by using a number of items and the total amount of the quantity of each item ordered.


Introduction
Data mining is an activity undertaken to explore the information-shaped pattern or grouping of data in a data set that has a large enough quantities to be taken of information that can be used to help make decisions. Data mining is also interpreted as an interesting pattern extraction of large amounts of data. In which a pattern is said to be interesting if the pattern is not implicit, previously unknown, but useful.
PT. Indomarco Prismatama is one of the largest franchised retail company in Indonesia. PT. Indomarco Prismatama have the space used to store merchandise while before they were distributed to retail stores called Distribution center. Distribution centers deliver goods based on data from the first order of goods sent by the store to the distribution center.
In the process of packing the goods to be delivered to the store frequently arise where frequent delays due to packing a large number of goods demand, but the number of personnel in the area are not qualified to do the packing at the request of such large items.
Based on the above problems, the researchers tried grouping goods order data with data mining. utilize goods order data sent by the store to find out what items are ordered by the store.

Data Mining
Data mining is the process to obtain useful information from large data base warehouse. Also defined as the extraction of new information, taken from a large a chunk of data that help in decision-making [1].
Data mining is the method used for extraction of hidden predictive information on the database [2]. The term data mining has several views, such as knowledge discovery or pattern recognition. the two terms actually has their accuracy, respectively. The term knowledge discovery or invention of appropriate knowledge for use purpose of data mining is to obtain knowledge that is still hidden inside a chunk of data. The term pattern recognition or pattern recognition also remains to be used because of the knowledge that was about to be extracted is shaped patterns that still need to be explored from a chunk of data that is facing [1].

K-Means Clustering
K-means clustering method is a method of group analysis that led to the object of observation of partitioning N into K groups (clusters) where each object observation is owned by a group with the closest mean [1]. K-means clustering can be interpreted as one of the data mining tools aimed at grouping the objects into clusters [2]. Grouping the data with K-means clustering algorithm is generally performed in the following order: a. Define k as the number of clusters to be formed. b. Initialize k centroid (center point cluster) beginning at random. c. Allocate any data or object to the nearest cluster. The distance between the object and the distance between objects in the cluster. specific data with the specified distance between the center of the cluster. To calculate the distance of all data to each cluster center, can use distances theory ecuilidean defined as the following equation.
Distance to the center of the cluster is recalculated with the current cluster membership. Center of the cluster is the average of all the data or objects in a particular cluster, if desired can also use the median value of the cluster. d. Repeat step three until the results of iteration is worth the same as the previous iteration.

Research methods
Research in an attempt to obtain the truth. research must be based on scientific thinking process as outlined in the scientific method.
Research framework is procedural used in conducting research studies in order to run smoothly and systematically with prior calculation phases or activities to be undertaken while doing this penelitian.Berikut is the framework of this study.

Analysis and Design
In this study, there are 19 data to be grouped. This data is the reservation data that have been through the cleaning process data in accordance with the stages of data mining.  Step 1: determine how many clusters to be formed. In this research will dibetuk 4 cluster.