Achmad Fauzi Bagus Firmansyah
Badan Pusat Statistik

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Kerangka Pemodelan Fase Tumbuh Padi Berbasis Machine Learning pada Wilayah dengan Pola Tanam Tidak Beraturan Maria Shawna Cinnamon Claire; Achmad Fauzi Bagus Firmansyah
Seminar Nasional Official Statistics Vol 2026 No 1 (2026): Seminar Nasional Official Statistics 2026
Publisher : Politeknik Statistika STIS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.34123/semnasoffstat.v2026i1.2807

Abstract

Rice phenology information is essential for agricultural statistics and food security policies. Although remote sensing-based machine learning has been widely applied for rice phenology estimation, studies integrating climate information in regions with irregular planting patterns while addressing operational scalability remain limited. This study develops an XGBoost- based framework that integrates Sentinel-1 data, rainfall information, and Area Sampling Frame (KSA) survey data as official-statistics-based ground truth in East Nusa Tenggara Province, Indonesia, where rice cultivation is predominantly rainfed. The results demonstrate that incorporating rainfall information improves model performance. Furthermore, a simplified modeling strategy that accounts for regional heterogeneity achieves high predictive performance without requiring separate models for each regional group, thereby reducing model development and maintenance complexity. These findings suggest that integrating multisource data, official statistics as ground truth, and a simplified modeling strategy provides a more accurate, efficient, and scalable framework for large-scale rice phenology estimation.