Claim Missing Document
Check
Articles

Found 1 Documents
Search

Prediksi Permintaan Keripik Buah dengan Metode Jaringan Syaraf Tiruan Backpropagation (Studi Kasus: CV. Arjuna 999) Benita Salsabila; Imam Cholissodin; Indriati Indriati
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 6 (2020): Juni 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The production of fruits in Indonesia tends to increase on a yearly basis. During harvest season, so much of those fruits would be left unsold or left to rot even though the sales value would also be lower than usual. Thus, a way to preserve the shelf-life of those fruits are needed so that they would not lose their value as quickly. One way to preserve fruits is to process them into dried fruit snacks, which is the expertise of CV. Arjuna 999 located in Batu, East Java. However, the process of turning real fruits into dried fruit snacks takes a while, which is why a strategy plan is needed to anticipate rising demands and the time it takes to make dried fruit snacks. The prediction uses an artificial neural network method, backpropagation. The dataset used contains of monthly dried fruit snacks demands of CV. Arjuna 999 starting from 2017 until 2019, with 80% of overall data used as training data while the other 20% is used as testing data. The result is a MAPE score of 4.429% which was derived from a combination of parameter values such as 10 (9 + 1 bias) hidden neurons, a learning rate value of 0.8 and a maximum iteration of 900.