PREDICTION OF U-DITCH RAW MATERIAL INVENTORIES IN. MULTIGUNA PRECAST MANDIRI USING A WEB-BASED SINGLE EXPONENTIAL SMOOTHING METHOD
No Thumbnail Available
Date
2020-12
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
PT. Multipurpose Precast Mandiri is a national company where the main activities are
service provision, heavy equipment supply, transportation, industry precast, printed
concrete, stone crusher. The purpose of this study is to use the method single exponential
smoothing to predict the availability of raw materials u-ditch in the coming months /
periods at PT. Multipurpose Precast Mandiri, designing and making raw material
inventory prediction applications u-ditch web based and measuring the level of accuracy
in predicting the availability of raw materials at PT. Multipurpose Precast Mandiri web
based. The method single exponential smoothing is a continuous activity to make
improvements in forecasting by averaging or smoothing the actual data values obtained
from past data by decreasing (exponential) with smoothing intervals between α (0.1 to 0,
9). Based on the research results, it is found that the prediction uses the method obtained
are the single exponential smoothing and the accuracy of the results is with the values
error MAD, MSE and MAPE, and the results prediction of the smallest sand raw material
at alpha 0.9 with an value of error MSE 276.9693789, MAD value 10.93720636 , and
MAPE 19.44665318% with good accuracy, the smallest cement raw material is at alpha
0.9 with an value of error MSE 162.0855996, MAD value 8.620003534, MAPE
18.08803119% with good accuracy, and the last is the smallest split stone raw material
at alpha 0.9 with value error MSE 264.4211558, MAD value 11.1994199, and MAPE
18.45811636% with good accuracy. A good level of accuracy is the level of accuracy that
gets good predictions with value error a small.
Description
Keywords
Forecasting, Forecasting, Inventory, Single Exponential Smoothing, MSE, Raw Materials, MSE, Raw Materials.