2026-11-03 –, Gardenia
This lightning talk introduces an early-stage approach to rice supply forecasting by combining rice-field polygon data, public statistics, and satellite-derived indicators such as vegetation and photosynthetic activity metrics.
This lightning talk presents an early-stage idea for forecasting rice supply conditions using open geospatial and statistical data. The approach combines rice-field polygon datasets, public agricultural statistics, and satellite-derived indicators related to vegetation condition, photosynthetic activity, and seasonal growth patterns.
Rather than focusing on a completed prediction model, the talk highlights how GIS boundaries, time-series satellite indicators such as NDVI-like vegetation measures, and public production data can be connected into a reproducible Python workflow. The goal is to explore how open data can support practical monitoring of rice-growing regions and provide early signals for agricultural supply analysis.
Takeo Shibata is an independent researcher and systems engineer with more than 30 years of experience in software and information systems development. He studied geophysics at Tohoku University and the University of California, Berkeley, and has maintained a long-standing interest in GIS, remote sensing, and Earth-related data analysis.
Alongside his professional work in the food distribution industry, he continues to explore the intersection of AI, geospatial technologies, and practical applications. He has presented topics related to AI and GIS at academic and technical communities, primarily in Japan and occasionally in the United States. His current interests include applying Python-based geospatial workflows and satellite data to real-world environmental and agricultural problems.