PREDIKSI BAHAN BAKU MAKANAN MENGGUNAKAN ALGORITMA LONG SHORT TERM MEMORY (LSTM) BERBASIS WEBSITE

Muhammad Farid Adika, . (2025) PREDIKSI BAHAN BAKU MAKANAN MENGGUNAKAN ALGORITMA LONG SHORT TERM MEMORY (LSTM) BERBASIS WEBSITE. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

[img] Text
ABSTRAK.pdf
Download (243kB)
[img] Text
AWAL.pdf
Download (791kB)
[img] Text
BAB 1.pdf
Restricted to Repository UPNVJ Only
Download (337kB)
[img] Text
BAB 2.pdf
Restricted to Repository UPNVJ Only
Download (546kB)
[img] Text
BAB 3.pdf
Restricted to Repository UPNVJ Only
Download (399kB)
[img] Text
BAB 4.pdf
Restricted to Repository UPNVJ Only
Download (1MB)
[img] Text
BAB 5.pdf
Download (253kB)
[img] Text
DAFTAR PUSTAKA.pdf
Download (229kB)
[img] Text
RIWAYAT HIDUP.pdf
Restricted to Repository UPNVJ Only
Download (100kB)
[img] Text
LAMPIRAN.pdf
Restricted to Repository UPNVJ Only
Download (2MB)
[img] Text
HASIL PLAGIARISME.pdf
Restricted to Repository staff only
Download (22MB)
[img] Text
ARTIKEL KI.pdf
Restricted to Repository staff only
Download (923kB)

Abstract

Food waste has become a major challenge in Indonesia, particularly due to the accumulation of unused food ingredients. This study aims to develop a prediction system for food ingredient requirements using the Long Short-Term Memory (LSTM) algorithm, integrated into a website. The dataset used consists of monthly sales data for two menu items from Warung Fotkop café, namely americano and garlic fries, covering the period from January 2021 to September 2024. The data underwent preprocessing through filtering, outlier removal, calculation of ingredient ratios, and normalization using RobustScaler. The LSTM model was evaluated using MSE, RMSE, MAE, and the R². For the Americano menu, the results showed an MSE of 26.7532, RMSE of 5.1724, MAE of 4.0826, and an R² score of 0.9963. Meanwhile, for the Garlic Fries menu, the model achieved an MSE of 16.1010, RMSE of 4.0126, MAE of 3.6161, and an R² score of 0.9835. These R² values, which fall within the range of 0.80 to 1.00, indicate that the model has a very strong level of accuracy in predicting monthly ingredient needs.. The model was then implemented as a website using Flask (backend) and ReactJS (frontend) with the Extreme Programming (XP) development approach. This website is capable of displaying predictions interactively and can assist store management in managing food stock more efficiently and sustainably. This research is expected to contribute to reducing food waste and supporting the principle of sustainability in the food industry.

Item Type: Thesis (Skripsi)
Additional Information: [No.Panggil: 2110511049] [Pembimbing 1: Indra Permana Solihin] [Pembimbing 2: Muhammad Adrezo] [Penguji 1: Ridwan Raafi'udin] [Penguji 2: I Wayan Rangga Pinastawa]
Uncontrolled Keywords: Food Ingredient Prediction, Long Short-Term Memory, sustainability, time series, website.
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Ilmu Komputer > Program Studi Informatika (S1)
Depositing User: MUHAMMAD FARID ADIKA
Date Deposited: 05 Aug 2025 06:51
Last Modified: 05 Aug 2025 06:51
URI: http://repository.upnvj.ac.id/id/eprint/37334

Actions (login required)

View Item View Item