Literatur Review Implementasi ArtificiaI Intellegence dalam Pertanian

Abstract

Abstract

The implementation of Artificial Intelligence (AI) in agriculture has become a focus of research to improve efficiency, productivity and sustainability of practices in the agricultural sector. Although much research has been conducted regarding the implementation of AI in agriculture, there is a significant gap in research in terms of focus on implementation potential, solutions to solve problems, and the algorithms and engineering methods used. This research uses the Systematic Literature Review (SLR) method by analyzing various studies on this topic. Data was collected from scientific journals, research reports, and other relevant publications. Literature results have answered the Research Question. Algorithms such as SVM, CNN, and LSTM are effective in detecting plant diseases, predicting crop yields, and optimizing resource use. ML, Computer Vision, Deep Learning, IoT, and Remote Sensing are also widely used in AI research in agriculture. This research found that SVM with IoT achieved 78.1% - 87.4% accuracy, remote sensing with machine learning achieved 81.5% - 99% for identifying land that needed additional pesticide and nutrient treatment, and AI integration with Big Data achieved 95% - 99% accuracy in detecting plant diseases and pest attacks. SVM also recorded an accuracy of 92.93% - 94.95% in soil type classification. These technologies are effective in increasing agricultural efficiency. AI in agriculture improves automatic irrigation with the SARIMAX Algorithm and QUHOMA platform, as well as detecting bad weather and crop diseases using drones and IoT. AI systems optimize crop types, planting times and precision pesticide applications with drones and robot sprayers to reduce environmental impact. Although the potential of AI in agriculture is enormous, further research is still needed with a focus on the algorithms and engineering methods used to maximize AI technology in the agricultural sector

Abstrak

Implementasi Artificial Intelligence (AI) di bidang pertanian telah menjadi fokus penelitian untuk meningkatkan efisiensi, produktivitas, dan keberlanjutan praktik di sektor pertanian. Meskipun banyak penelitian telah dilakukan mengenai implementasi AI di bidang pertanian, terdapat kesenjangan dalam penelitian yang signifikan dalam hal fokus pada potensi implementasi, solusi untuk menyelesaikan masalah, serta algoritma dan metode teknik yang digunakan. Penelitian ini menggunakan metode Systematic Literature Review (SLR) dengan menganalisis berbagai penelitian mengenai topik tersebut. Data dikumpulkan dari jurnal ilmiah, laporan penelitian, dan publikasi terkait lainnya yang relevan Hasil Literatur telah menjawab Research Question. Algoritma seperti SVM, CNN, dan LSTM efektif dalam mendeteksi penyakit tanaman, memprediksi hasil panen, dan mengoptimalkan penggunaan sumber daya. ML, Computer Vision, Deep Learning, IoT, dan Remote Sensing juga digunakan luas dalam penelitian AI di pertanian. Penelitian ini menemukan bahwa SVM dengan IoT mencapai akurasi 78.1% - 87.4%, remote sensing dengan machine learning mencapai 81.5% - 99% untuk identifikasi lahan yang membutuhkan perlakuan pestisida dan nutrisi tambahan, dan integrasi AI dengan Big Data mencapai akurasi 95% - 99% dalam mendeteksi penyakit tanaman dan serangan hama. SVM juga mencatat akurasi 92.93% - 94.95% dalam klasifikasi jenis tanah. Teknologi-teknologi ini efektif meningkatkan efisiensi pertanian. AI dalam pertanian memperbaiki irigasi otomatis dengan Algoritma SARIMAX dan platform QUHOMA, serta mendeteksi cuaca buruk dan penyakit tanaman menggunakan drone dan IoT. Sistem AI mengoptimalkan jenis tanaman, waktu tanam, dan aplikasi pestisida presisi dengan drone dan robot penyemprot untuk mengurangi dampak lingkungan. Meskipun potensi AI dalam pertanian sangat besar, penelitian lebih lanjut masih diperlukan dengan fokus pada algoritma dan metode teknik yang digunakan untuk memaksimalkan teknologi AI di sektor pertanian.