SISTEM MONITORING KEKERINGAN BERDASARKAN NORMALIZED DIFFERENCE VEGETATION INDEX (NDVI) UNTUK PREDIKSI PRODUKSI LAHAN PADI MENGGUNAKAN ALGORITMA RANDOM FOREST (STUDI KASUS: GOMBONG, KEBUMEN, JAWA TENGAH)

RAHMANA, IRZA HARTIANTIO (2021) SISTEM MONITORING KEKERINGAN BERDASARKAN NORMALIZED DIFFERENCE VEGETATION INDEX (NDVI) UNTUK PREDIKSI PRODUKSI LAHAN PADI MENGGUNAKAN ALGORITMA RANDOM FOREST (STUDI KASUS: GOMBONG, KEBUMEN, JAWA TENGAH). S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (HAL COVER) 01 COVER.pdf Download (252kB) | Preview
	Text (BAB I) 02 BAB 1.pdf Restricted to Registered users only Download (215kB)
	Text (BAB II) 03 BAB 2.pdf Restricted to Registered users only Download (693kB)
	Text (BAB III) 04 BAB 3.pdf Restricted to Registered users only Download (244kB)
	Text (BAB IV) 05 BAB 4.pdf Restricted to Registered users only Download (947kB)
	Text (BAB V) 06 BAB 5.pdf Restricted to Registered users only Download (79kB)
	Text (DAFTAR PUTAKA) 08 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (260kB)
	Text (LAMPIRAN) 09 LAMPIRAN.pdf Restricted to Registered users only Download (573kB)

Abstract

Rice is one of the popular carbohydrate sources and is in demand by some Indonesian people, such as corn, sago, sugar palm, cassava, sweet potato, or taro. The distribution of rice plants is evenly distributed in almost all parts of Indonesia. Kebumen Regency is one of the buffers for rice production in Central Java Province. The agricultural sector is constantly faced with high risk and uncertainty, and this comes from natural environmental factors such as natural disasters (floods and droughts). Currently, the threat of drought has hit some areas in Kebumen Regency. Farmers and the district government are looking for various alternative water sources to drain it using a pumping system. There needs to be a systematic and efficient effort with the lowest risk of loss due to the threat. The purpose of this study was to determine the implementation of the Random Forest Regression algorithm for predicting rice production in Gombong District, Kebumen District, Central Java Province, making a prediction system for rice production in Gombong District, Kebumen District, Central Java Province using the Random Forest Regression algorithm, and recommendations. The best way to increase rice production in Gombong District, Kebumen Regency, Central Java Province. The benefit of this research is to build a predictive system for rice production in monitoring drought in Gombong District, Kebumen Regency, Central Java Province using the Random Forest Regression algorithm. The prediction system for rice production in Gombong District, Kebumen Regency, Central Java Province uses the Normalized Difference Vegetation Index (NDVI) and the Random Forest Regression algorithm. The evaluation uses several scoring methods such as Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), R², Explained Variance Score (EVS), and Out-of-Bag (OOB). This research shows that from March 1, 2020, to July 31, 2020, the average MAE is 0.0150064182730663, MSE is 0.0002517777048969, RMSE is 0.0151865750792201, R² is 0.9940160450705040, EVS is 0.9935376833643790, and OOB is 0.9869360548888210 so that it proves that in the period of rice growth in the vegetative phase (0-60 days) there is drought. Keywords: prediction system, NDVI, random forest, regression, rice Padi merupakan salah satu sumber karbohidrat yang populer dan diminati oleh sebagian masyarakat Indonesia, seperti halnya jagung, sagu, aren, singkong, ketela rambat, atau talas. Penyebaran tanaman padi merata hampir di seluruh wilayah Indonesia. Kabupaten Kebumen merupakan salah satu penyangga produksi padi di Provinsi Jawa Tengah. Sektor pertanian selalu dihadapkan pada sebuah risiko serta ketidakpastian yang tinggi, hal tersebut berasal dari faktor lingkungan alam seperti bencana alam (banjir dan kekeringan). Ancaman kekeringan pada dewasa ini telah melanda sebagian wilayah di Kabupaten Kebumen. Petani bersama pemerintah kabupaten mencari berbagai sumber air alternatif hingga mengalirkannya menggunakan sistem pompanisasi. Perlu adanya suatu upaya sistematis dan efisien yang memiliki risiko kerugian paling rendah akibat ancaman tersebut. Tujuan penelitian ini adalah mengetahui implementasi algoritma Random Forest Regression untuk prediksi produksi lahan padi di Kecamatan Gombong, Kabupaten Kebumen, Provinsi Jawa Tengah, pembuatan sistem prediksi produksi lahan padi di Kecamatan Gombong, Kabupaten Kebumen, Provinsi Jawa Tengah menggunakan algoritma Random Forest Regression dan rekomendasi terbaik untuk meningkatkan produksi lahan padi di Kecamatan Gombong, Kabupaten Kebumen, Provinsi Jawa Tengah. Manfaat dari penelitian yang dilakukan ialah membangun sistem prediksi produksi lahan padi dalam monitoring kekeringan di Kecamatan Gombong, Kabupaten Kebumen, Provinsi Jawa Tengah menggunakan algoritma Random Forest Regression. Sistem prediksi produksi lahan padi di Kecamatan Gombong, Kabupaten Kebumen, Provinsi Jawa Tengah ini menggunakan Normalized Difference Vegetation Index (NDVI) dan algoritma Random Forest Regression. Untuk evaluasi menggunakan beberapa scoring seperti Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), R², Explained Variance Score (EVS) dan Out-of-Bag (OOB). Penelitian ini menunjukkan pada 1 Maret 2020 hingga 31 Juli 2020 rata-rata MAE sebesar 0,0150064182730663, MSE sebesar 0,0002517777048969, RMSE sebesar 0,0151865750792201, R² sebesar 0,9940160450705040, EVS sebesar 0,9935376833643790 dan OOB sebesar 0,9869360548888210. Sehingga pada membuktikan bahwa pada periode pertumbuhan padi fase vegetatif (0-60 hari) terdapat kekeringan. Kata kunci: sistem prediksi, NDVI, random forest, regression, padi

Actions (login required)

View Item