Wanda Yuliana Sri Utami, - (2024) ANALISIS PERBANDINGAN PERFORMA ALGORITMA SARIMA DAN LSTM DALAM MEMPREDIKSI PENJUALAN PRODUK FASHION ANAK (STUDI KASUS CEELIK KIDS APPAREL). S1 thesis, Universitas Pendidikan Indonesia.
![]() |
Text
S_RPL_2009164_Title.pdf Download (629kB) |
![]() |
Text
S_RPL_2009164_Chapter1.pdf Download (176kB) |
![]() |
Text
S_RPL_2009164_Chapter2.pdf Restricted to Staf Perpustakaan Download (207kB) |
![]() |
Text
S_RPL_2009164_Chapter3.pdf Download (207kB) |
![]() |
Text
S_RPL_2009164_Chapter4.pdf Restricted to Staf Perpustakaan Download (793kB) |
![]() |
Text
S_RPL_2009164_Chapter5.pdf Download (37kB) |
![]() |
Text
S_RPL_2009164_Appendix.pdf Restricted to Staf Perpustakaan Download (134kB) |
Abstract
Dalam industri fashion anak yang dinamis, fluktuasi pola pembelian memerlukan prediksi yang akurat. Ceelik Kids Apparel mengalami masalah dengan metode prediksi kasar yang sering mengakibatkan kelebihan atau kekurangan stok. Penelitian ini bertujuan untuk membandingkan performa dua metode prediksi untuk menentukan algoritma dengan akurasi tertinggi dan memberikan panduan dalam memilih metode yang efektif. Metode statistika yaitu dengan algoritma SARIMA (Seasonal Autoregressive Integrated Moving Average) dan metode deep learning yaitu dengan algoritma LSTM (Long Short-Term Memory). SARIMA adalah metode statistika yang efektif dalam menangani pola musiman dalam data deret waktu, sedangkan LSTM adalah pendekatan deep learning yang dapat menangkap pola kompleks dalam data. Penelitian ini menggunakan data penjualan historis dari Ceelik Kids Apparel, membagi data menjadi set pelatihan dan set pengujian untuk evaluasi performa masing-masing algoritma dan dilakukan prediksi 1 tahun kedepan. Hasil penelitian menunjukkan bahwa SARIMA memberikan tingkat akurasi yang sangat baik dengan Mean Absolute Percentage Error (MAPE) sebesar 4.1%. Sebaliknya, LSTM menghasilkan MAPE sebesar 17.4%, yang masih dalam kategori "kemampuan prediksi baik" tetapi tidak seakurat SARIMA. Secara keseluruhan, algoritma berbasis statistika seperti SARIMA menunjukkan hasil yang lebih baik dibandingkan dengan pendekatan deep learning seperti LSTM dalam prediksi penjualan produk fashion anak. ---------- In the dynamic children's fashion industry, fluctuating purchasing patterns require accurate forecasting. Ceelik Kids Apparel faces issues with rough forecasting methods that often lead to either excess stock or stock shortages. This study aims to compare the performance of two forecasting methods to determine the algorithm with the highest accuracy and provide guidance on selecting the most effective method. The statistical method used is the SARIMA (Seasonal Autoregressive Integrated Moving Average) algorithm, while the deep learning method is the LSTM (Long Short-Term Memory) algorithm. SARIMA is effective in handling seasonal patterns in time series data, whereas LSTM is a deep learning approach capable of capturing complex patterns in the data. This study uses historical sales data from Ceelik Kids Apparel, splitting the data into training and testing sets to evaluate the performance of each algorithm and making predictions for the next year. The results show that SARIMA provides very good accuracy with a Mean Absolute Percentage Error (MAPE) of 4.1%. In contrast, LSTM yields a MAPE of 17.4%, which is still in the "good prediction ability" category but not as accurate as SARIMA. Overall, statistical algorithms like SARIMA perform better than deep learning approaches like LSTM in forecasting children's fashion product sales.
Item Type: | Thesis (S1) |
---|---|
Additional Information: | https://scholar.google.com/citations?hl=id&user=6_7NyW4AAAAJ SINTA ID : 6658552 SINTA ID : 6681751 |
Uncontrolled Keywords: | Prediksi, Statistika, Deep Learning, SARIMA, LSTM, Forecasting, Statistical |
Subjects: | L Education > L Education (General) L Education > LB Theory and practice of education L Education > LB Theory and practice of education > LB1501 Primary Education Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) |
Divisions: | UPI Kampus cibiru > S1 Rekayasa Perangkaat Lunak |
Depositing User: | Wanda Yuliana Sri Utami |
Date Deposited: | 23 Aug 2024 02:48 |
Last Modified: | 23 Aug 2024 02:48 |
URI: | http://repository.upi.edu/id/eprint/119790 |
Actions (login required)
![]() |
View Item |