IMPLEMENTATION OF SES TECHNIQUES IN THE NEEDS FOR BROILER CHICKEN MEAT IN ASAHAN DISTRICT

: The Asahan Regency Animal Husbandry and Animal Health Service carries out outreach activities on the level of demand for boiler chicken, which means that the department must be more active in collecting data on the volume of boiler chicken every year because consumer demand for chicken continues to increase and instability often occurs. Where in the period 2012 to 2024 there was a significant increase in demand. So it becomes an obstacle in the stock of chickens in the Asahan Regency market which must be prepared. If there is too much chicken supply but the demand for chicken consumption decreases then this will result in a loss, conversely if the demand for chicken consumption increases but the chicken supply cannot be prepared then this will be a loss for chicken traders and distributors. To increase the accuracy of the estimates, the use of the Single Exponential Smoothing Method is proposed. This method smooths past values with exponential weights, placing greater emphasis on recent data. Using data from the last 12 (twelve) years as reference data for recording the past for forecasting experiments for the next 1 (one) month. Experiments using different Alphas test accuracy in actual situations. The results of the experiment with an alpha of 0.9 are almost the same as the actual value in predicting the situation one year into the future with the lowest error, namely MAPE 7.4% and forecasting broiler chicken meat supplies of 13661261.46 chickens.


INTRODUCTION
The increase in population and improvements in people's living standards have caused demand for various food needs to continue to increase.
ncreasing consumption of animal protein in line with public awareness of nutrition.Based on BPS Susenas 2022 data, Indonesia's largest consumption of animal protein is specifically Asahan Regency which requires broiler chicken meat.
Dinas Kesehatan dan Peternakan Kabupaten Asahan carries out outreach activities on the level of demand for broiler chicken, which means that the department must be more active in collecting data on the volume of broiler chicken every year because consumer demand for chicken continues to increase.
So far, the department has only applied a very simple calculation method, namely predicting demand for chicken which always changes every year using data from 2012 to 2023.This will make it easier for related parties, especially the Asahan Regency Animal Husbandry and Health Service, to quickly make the right decisions in predicting demand for chickens in the future, so that they can find solutions to meet increasing consumer demand or excess stock, therefore a system is needed Forecasting is a calculation analysis technique carried out using approach to estimate future events using references to past data to minimize the influence of uncertainty.Forecasting itself can be short, medium or long term for a company.Forecasting is a very important tool in effective and efficient planning [1].
From previous research including the title " Single Exponential Smoothing: Method for Forecasting Vaccine Needs Measles" [2], The analysis technique for system forecasting uses the SES method, while the system testing uses a Blackbox Our findings show that the lowest MAPE value was obtained at 49.8%.The results of testing the system using a Blackbox that all components in this system are already functioning properly.With this system, it can make it easier for related parties to predict the number of measles vaccines in the new Gambir health center.
Comparative Analysis of Single and Single Exponential Smoothing Methods Moving Average in Order Forecasting [3], Forecasting using the time series method, namely SES and SMA, the forecasting model is obtained from the actual data used in this case study, the forecasting model using the SMA method is declared very good because it has a MAPE value <10% while the SES method is declared good because it has a MAPE value.= 11% (10-20%).The SMA method has an error rate that is more sophisticated than the SES method.For forecasting order requests in this study, the SMA method is better used because it has the smallest error rate in forecasting.A solution is needed in this case form a capable forecasting system to predict future demand for broiler chicken meat, which Later it will become a reference in the process of restocking goods.Therefore, single exponential smoothing method is used to forecast the amount of demand for broiler chicken meat, especially when the data shows seasonal trends and patterns.By applying this method in information systems, those it is hoped that forecasting can be carried out with minimal and approximate error rates actual value, so that the Asahan District Animal Husbandry and Animal Health Service can Estimate the amount of broiler chicken meat for the next time period and avoid losses

METHOD
Quantitative methods use various mathematical models that rely on historical data to estimate demand for broiler chicken meat.The framework stages carried out as follows: Image 1. Research Framework

Identify The Problem
The problems identified in this research.There is a demand for broiler chicken meat each year that cannot be predicted by the Asahan District Animal Husbandry and Animal Health Service.It is difficult to determine the need for broiler chicken meat, so it does not match the amount of supply demanded by consumers.There is no forecasting system for the need for broiler chicken meat, resulting in a reduced supply of building materials.

Literature Study or Library Study
After identifying the problem, the carried out a literature study by searching for theoretical foundations obtained from various sources to complete the concepts and theories, so that they have a good and appropriate scientific basis.

Data Collection
The data collection stage was carried out after literature study and problem identification.Data collection is a technique or method used by to collect data.Data collection is carried out in order to obtain the information needed to achieve research objectives.Meanwhile, data collection instruments are tools used to collect data.Because it is a tool, the data collection instruments can be interviews and observations.The data taken is data on the amount of meat needed for broiler chickens by taking data from the previous period.

System Analysis
After collecting data, the made a system analysis aimed at knowing the forecasting system for the need for broiler chicken meat.Following are the problems that occurred.The Asahan Regency Livestock and Animal Health Service cannot yet predict demand for broiler chicken meat in the next period.The Asahan District Animal Husbandry and Animal Health Service often experiences excess and shortage of broiler chicken meat stocks.For this reason, an information system was created, namely forecasting the need for broiler chicken meat using the Exponential Smoothing method at the Asahan District Livestock and Health Service.

System Design
This stage was carried out after the carried out a system analysis and identified the obstacles and problems that occurred.System design an activity of designing and processing an information system from the results of system analysis so that it can meet the needs of users.

System Implementation
The next stage namely the implementation stage where at this stage the ses method is applied to help The Single Exponential Smoothing (SES) method is a method that uses very little past data recording and assumes fluctuating or unstable data.Exponential smoothing is a weighted moving average forecasting technique where data is weighted by an exponential function [11], Exponential smoothing a moving average forecasting method with sophisticated weighting, but is still easy to use [12].
The single exponential smoothing (SES) method a procedure by repeating calculations continuously using the latest observation data, each data used in this method is given a weight which is symbolized  (alpha) alpha value which ranges from 0 to 1 which produces a level value that error the smallest will be selected for use in the forecasting model [13]  The results of forecasting Broiler Chicken can be seen from the Single Exponential Smoothing method using Alpha α 0.1 to 0.9 as follows: Sample testing used (0.1 to 0.9), the smallest error value was obtained using 0.9 which obtained an error value of 3.09% with forecasting results for the 2024 period, namely 13661261.46chickens for broiler meat needs.After the forecasting results are obtained, this can be something that helps consider the estimated decision in determining the quota for boiler chicken meat

RESULT AND DISCUSSION
The implementation of forecasting demand for broiler chickens at the Asahan Regency Animal Husbandry and Health Service using the Single Exponential Smoothing (SES) method was carried out using the PHP programming language and the database used was MySQL.

Main Menu Page
If the user had been successfully logged in, the main menu page of the broiler meat demand forecasting system for the Asahan Regency Animal Husbandry and Health Service will appear.

Input Data On The Type Of Demand
Next, the form for the type of meat requirement displayed in the broiler meat demand forecasting system at the Asahan Regency Livestock and Health Service.There are several types of broiler chickens obtained from various regions, they need to be input into the system.The broiler chicken request type input system consists of a field containing the code and type of chicken as follows:

Table 2 .
Analysis of Forecasting Errors

Table 3 .
Results of Error Forecasting