Anticipating the Scarcity of Food Commodity to Sufficient Availability for the Community in East Java with a Comparison Approach to Forecasting Artificial Neural Networks and Exponential Smoothing

Abstract

This research was conducted to predict food commodity prices which include beef prices, medium quality rice price data, chicken egg prices, sugar prices, branded cooking oil prices, bulk cooking oil prices, curly red chili prices, garlic prices, shallot prices and price of grilled chicken the research carried out has a contribution to the commodity variables which are added as part of the factors causing food shortages in 2021, whereas in research related to predictions as was done by previous researchers that variables related to activity objects become a reference in the prediction process. This research was conducted by analyzing food commodity price data taken from the weekly period, namely December 2020 (IV)-April 2022 (I) in East Java Province using the exponential smoothing method. From the results of the analysis it is known that the smallest MAPE (Mean Absolute Percentage Error) ranges from 0.22%-2.44%, which means that the prediction accuracy of food ingredients using the triple exponential smoothing method is very high, which is greater than 97.66% so that it can be used as a recommendations in forecasting food commodity prices for the next period.