Analisis Metode Backpropagation Dalam Memprediksi Pembungkusan Bunga Pada Pohon Induk Kelapa Sawit
Abstract
The oil palm tree is a key crop in the palm oil industry, considering the importance of a key stage in oil palm fruit production, namely flower packing on the mother tree. Determining the optimal timing for the flower packing process has a significant impact on yield and palm oil quality. Currently, the Oil Palm Research Center (PPKS) not only conducts research related to oil palm, but also provides quality seeds, hence the need for flower wrapping techniques to be used in research and production of oil palm seeds to ensure the desired quality and characteristics of the plants that will grow from these seeds. Therefore, predicting the optimal time for flower packing becomes very important in an effort to increase production efficiency and yield. The backpropagation method is one of the commonly used techniques in artificial neural networks to predict and classify data. In this study, data on flower packing on oil palm mother trees were collected and organized into a suitable dataset. Then, an artificial neural network model with backpropagation architecture is implemented and trained using the dataset. Evaluation and analysis were conducted on the performance of the model in predicting flower packing using the backpropagation method. The training data used is from 2011 to 2021, while the testing data is from 2012 to 2022. The method used is the backpropagation algorithm, there are 5 network architectures implemented in the Matlab application. The architectures used include 10-88-1 with a Mean Squared Error of 0.000278773, 10-63-1 with a Mean Squared Error of 0.000247101, 10-74-1 with a Mean Squared Error of 0.000289653, and 10-79-1 with a Mean Squared Error of 0.000171904. Based on the research results, the best architecture produced is 10-36-1 with an accuracy of 92% and a test Mean Squared Error of 0.000100008 at epoch 25414 Iterations. Thus it can be concluded that the backpropagation algorithm can provide good accuracy in the prediction process.
References
[2] R. F. P. Sinaga, B. D. Setiawan, dan Marji, “Peramalan Produksi Kelapa Sawit Menggunakan Jaringan Syaraf Tiruan Dengan Metode Backpropagation (Studi Kasus PT . Sandabi Indah Lestari),” J. Pengemb. Teknol. Inf. dan Ilmu Komput., vol. 2, no. 11, hal. 4613–4620, 2018.
[3] F. K. Harahap, “Pengadaan bahan tanaman kelapa sawit”, Laporan Praktik Kerja Lapangan, Yogyakarta, Politeknik Lpp 2021.
[4] R. Aulia, “Penerapan Metode Backpropagation Untuk Memprediksi Jumlah Kunjungan Wisatawan Berdasarkan Tingkat Hunian Hotel,” Jurteksi, vol. 4, no. 2, hal. 115–122, 2018, doi: 10.33330/jurteksi.v4i2.45.
[5] O. Armaya Putri, H. Satria Tambunan, S. Tunas Bangsa, A. Tunas Bangsa, dan I. A. Jln Sudirman Blok No, “Prediksi Kunjungan Wisatawan Mancanegara Ke Indonesia Menggunakan Jaringan Saraf Tiruan Dengan Algoritma Backpropagation,” Januari, vol. 2, no. 1, hal. 1–7, 2021.
[6] M. Birky, A. Akbar, dan A. A. Supianto, “Optimasi Peramalan Metode Backpropagation Menggunakan Algoritme Genetika pada Jumlah Penumpang Kereta Api di Indonesia,” J. Pengemb. Teknol. Inf. dan Ilmu Komput., vol. 3, no. 3, hal. 2533–2541, 2019.
[7] B. Yanto, Hendri, Almadison, R. Hutagaol, dan R. Rahman, “Analisis Optimasi Algoritma Backpropagation Momentum Dalam Memprediksi Jenis Tingkat Kejahatan Di Kecamatan Tambusai Utara,” J. Ict Apl. Syst., vol. 1, no. 1, hal. 47–60, 2022, doi: 10.56313/jictas.v1i1.165.