Forecasting analysis at PT. lion metal works using artificial neural network

Abstract

PT. Lion Metal Works is one of the manufacture company which produces office furnitures. PT. Lion Metal Works has problem in determining demand of products because they do not do the forecasting method. The aim of this research is to make an application program in order to help the company do the forecasting calculation easily by using artificial neural network as the forecasting technique to resolve shortage and over production problem. Data needed to do calculation by using ANN is demand data from January 2017 until December 2019. Verification forecasting is needed to check whether the forecasting function is already representing exiting data or not, for the example result of L-31K MAD 0,0025, MSE 0,001 SDE 0,0034, MPE 0,0014, and MAPE 0,0037. Result of validation between ± 4 and prove the calculation is valid. is After finish doing MATLAB calculation the output is normalization data and its need to do denormalization data to getting unit product data for each item.