Volume 1 Number 2 Juli-December 2021 PP 77-80 INTELMATICS ISSN 2775-8850 *Corresponding author E-mail address: jeanyfadhilah@gmail.com 77 D Designing Data Warehouse For Forecast and Data Visualization of Sales Nutrition Products Jeany Fadhilah Agatha Siahaan 1* , Dedy Sugiarto 2 , Teddy Siswanto 3 Study Program Information System 123 , Faculty of Industrial Technology, Trisakti University Abstract—Sales data can be processed in such a way that it can become information that is used as material for analysis and consideration in making decisions. This study aims to visualize PT XYZ sales data for nutritious intake products and predict sales figures for 2018 and 2019. Data is obtained directly from PT XYZ by submitting a request for data withdrawal. Data on sales of nutritious beverage products for the last 5 years are processed using Pentaho tools with ETL method (extract, transform, load) then predicted sales figures for 2019 using R programming language with ARIMA and Holt-Winters methods after which data will be visualized using Powe BI so that the display of data presentation is more interesting and informative. To find out the compatibility in using the forecasting method, the writer will compare RSME numbers from both methods and use the method with the smallest RSME number. Keywords: Forecasting, Pentaho, ETL (Extact, Transform, Load), Data Visualization, ARIMA, Holt-Winters, RSME I. INTRODUCTION ata is a value that contains details of an object or event. Data can be in the form of symbols, numbers, characters, images, sounds, or other signs. Data that has not been processed is still irregular so that the meaning is unclear and difficult for the recipient to understand. Turning data into information, information into knowledge, and knowledge into plans will trigger actions that are beneficial to businesses in accordance with the understanding of business intelligence. The company has a lot of data for which proper data processing is needed. Like PT. XYZ who has sales data and wants to process it into useful information in making decisions and designing strategies. Sales data can be processed into OLAP data warehouse using Kimball's 4 Steps namely determining business processes, identifying grains, identifying dimension tables, and finally identifying fact tables. Sales data can also be processed to predict sales figures can also be changed the presentation of data by visualizing data. Data visualization is a modern form of communication to convey information by representing it in graphic form. Pictures are easier to understand than writing. In previous studies forecasting using the Holt- Winters and ARIMA methods conducted by Tias Safitri to compare the best method for predicting the number of foreign tourist arrivals to Bali Ngurah Rai in 2010 to 2015 [1]. In this study the authors made predictions using the Holt-Winters and ARIMA methods while for the visualization of the data the authors used power BI as a tool. II. THEORY Data Warehouse is a system that extracts, cleans, adjusts, and transmits source data into dimensional data storage then supports and implements requests and analyzes for decision making purposes [2]. ETL (Extract, transform and load) is a sequence of applications starting from extracting a collection of data from various sources, then bringing it to the data staging area, and applying the sequence of the process of preparing data to be migrated to the data warehouse [3]. Spoon / Kettle is an application that functions to process ETL (Extract, Transformation and Load) in Business intelligence. Besides the enterprise version, there is a community version that can be used for free. Pentaho Data Integration can be run as a standalone application or as a client-server app, where development is done on a computer and execution is executed on the server[4]. ARIMA is one of the predictions in statistical techniques that use time series in predicting data in the future. The parameters used in ARIMA are p, d, q which refer to the autoregressive, integrated and moving average portion of the dataset. The ARIMA prediction technique will handle trends, seasons, cycles, errors, and non-stationary aspects of the data set when making predictions [5]. Holt-Winter Python is one type of exponential smoothing where the time series is described with an increasing or decreasing trend or seasonality. Exponential Smoothing Holt- Winterss can also be used to make short-term predictions [5]. III. METHODS The initial step of the method used for this forecasting is the collection of raw data by sending an official data withdrawal request letter from the university. Raw data on sales of PT XYZ nutritional beverage products taken in the form of a Microsoft Excel spreadsheet file. Next, design the data warehouse using the 4 steps methodology of Kimball Kimball's 4 steps methodology consists of: (1) determinin