Analysis Forecasting Sales with Single Exponential Smoothing Method

- Currently, as we know that the use of computer-based systems, almost everyone's activities depend on computer-based systems. one that is currently very adequate in the use of the application is the sales system, in this research the problem faced by the company is the difficulty of determining the amount of inventory stock each month. The reason is because the number of types of goods is not small so this is the cause of the lack of fulfillment of consumer demand every day. Companies that do have a number of inventories whose stock or inventory is on a large scale, so it needs a special system in this control so that the required inventory is always met, the point is for the company's income to increase even more. Researchers as technology-savvy people provide ideas for companies to use a system that can minimize the lack of goods inventory. This is what the researcher wants to solve the problem with the single exponential smoothing method. Because the method used can do fortune-telling in the short term for example 1 month ahead. The results of this study succeeded in predicting the amount of inventory in stock every month.


INTRODUCTION
Stock adjustment of inventory data is one of the essential parts of every company engaged in sales. Inventory goods is one of the most critical merchandise in a trading company (Ferawati et al., 2020) In every business, stock of goods is one thing that is very important for the smooth sale of goods to customers (Zalukhu & Handriani, 2019) nd they are also used for production continuity (Pradana & Jakaria, 2020). High inventory allows the company to meet sudden demand in sufficient quantities, evenly distributed and affordable to the community (Sukanda & Dewi, 2018).
Every company engaged in sales is constantly experiencing an increase and decrease in transactions. Each officer must provide a system that can conduct sales transactions and inventory processing effectively and efficiently. So far, most companies still use a manual system whose work must take a long time to forecast the stock of goods available by sales. Each officer concerned with the inventory of goods is required to provide a plan every month or every week so that can be ordered from the supplier according to the stock of goods that have run out. Research carried out by an information system becomes a tool for forecasting in more effective and efficient planning with the exponential smoothing method (Raharja et al., 2013). The Exponential Smoothing method is also called Exponential smoothing, which is one of the forecasts with an average taking technique because the weighting is given by an exponential function (Raharja et al., 2013), (Fachrurrazi, 2015), (Faisol & Aisah, 2016). Exponential Smoothing is a fairly good forecasting method in the long and medium-term (Faisol & Aisah, 2016). The exponential smoothing method parameter is denoted by α (alpha). This picture is used to process the employee section because, with this method, pharmacies can predict what steps will be taken to meet consumer demand. Forecasting or interpreting the results obtained cannot provide 100% results in forecasting data because the future is uncertain. What is uncertain is what can be tested using different alphas to get maximum results. In this study, we succeeded in carrying out sales forecasting based on the transaction history in the past months, which is reference data in forecasting (Al'afi et al., 2020). Previous researchers said that forecasting predicts future events by utilizing past data (Prasetya et al., 2015). Forecasting aims to minimize forecasting errors that can be measured using Mean Absolute Deviation (MAD) and Mean Square Error (MSE). The other purpose of this forecasting is to predict future events and take advantage of past data (Fachrurrazi, 2015). Time horizon in Forecasting is usually classified based on future time, and Forecasting is an attempt to forecast future situations with data from the past. Based on the time horizon, Forecasting can be divided into three types : A. Short range forecast It is typically less than 3 months but has a time span of up-to 1 year. It is used in planning, purchasing for job schedules, job assignments, work force levels, product levels. B. Medium range forecast It is typically 3 months to 1 year but has a time span from one to three years. It is used for sales planning, production planning, cash budgeting and so on. C. Long range forecast This has a time span of three or more years. It is used for designing and installing new plants, facility location, capital expenditures, research and development, etc. Sales predictions are not just forecasting or simple additions, but this is one that can control inventory so that the stock needed by consumers is still fulfilled (Gustriansyah, 2017). Information is a fact that has been processed in a certain way that describes a real event to be processed so that it can be understood and used in making a decision (d. Y. K. Harfizar, 2017).

RESEARCH METHODOLOGY
In the opinion of Heizer and Rander (2015, 118), to make a demand forecast must use a particular method. All forecasting methods have the same idea, to use past data to predict or project future data. Based on the technique, forecasting methods can be categorized into qualitative and quantitative methods. The dependent variable to perform forecasting will remain the same, namely, and the independent variable is x. Based on previous research, forecasting must use specific methods, even though the forecasting method has a good way of processing data, that is, using past data to predict future data. The technique used is the forecasting method, namely qualitative and quantitative.

Mulai
Description : Ft = the current basic value for the current period (t) At-1 = actual demand for the current period (t) Ft-1 = the provious basic value from the previous period α = smoothing factor for the basic value Paradigma, Vol. 24, No. 2, September 2022P-ISSN 1410-5063, E-ISSN: 2579  In the table above is the data that will be used in this study, where the data to be used is from January 2021 to December 2022 The method that forecasting will use is to solve the problem with the single exponential smoothing method. This forecasting method will

CONCLUSION
The sale of basic necessities is part of the distribution of the sale and purchase of basic food categories. The demand for goods is a point in the distribution of customers for goods. Employees act to prepare and record purchase orders for goods for the following month. The activity of sorting available and unavailable inventory data often takes a long time. The Exponential Smoothing method is a forecasting method on moving averages by giving weights that are easy to analyze. Time series sales data with Exponential Smoothing forecasting method is expected to be able to handle optimal inventory for inventory control. The use of data from the last 2 years as reference data for past recordings for forecasting experiments for the next 3 months. Experiments with different Alphas tested previously and previously used data from the previous 6 months in research using data from the last 2 years. This system can provide recommendations in the stock of goods so that the supply of sales is still met.