Open AccessThe Simple Boxplot Method for an Effective Prediction
Author(s) -
Reonaldo,
Manatap Dolok Lauro,
Dyah Erny Herwindiati
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1007/1/012141
Subject(s) - outlier , mean absolute percentage error , statistics , computer science , data mining , test data , mean squared error , mathematics , programming language
This study aims to do a prediction of demand goods at a factory for 1 day ahead using double moving average method and comparing the forecasting results. Data source come from two different types of data which are complete data and clean data. Clean data was an optimal data that has been cleaned from outlier using boxplot method. The data source used in the calculation is simulation data for 945 days. Based on the test results, Shows the results of forecasting using complete data that is equal to 4692 with MAPE 6.88 while the results of forecasting use clean data that is equal to 4876 with MAPE 3.84. From these results, it can be concluded that forecasting using clean data is more accurate than forecasting using complete data because the smaller the error rate (MAPE) produced, the better the accuracy.