
Predicting crop yields is a significant challenge in modern agriculture, especially under fluctuating environmental conditions. With the escalating demand for sustainable food production, understanding the complex interplay between environmental factors and crop growth has become increasingly crucial. These challenges underscore the need for in-depth research on the interactions between crop yields and environmental variables over time.
Conducted by a team from Shiga University and published in Horticulture Research on May 24, 2024, the study presents a novel functional data analysis (FDA)-based methodology for predicting crop yields in year-round cultivation. By analyzing time-series data of strawberries and tomatoes grown in natural-light plant factories, the research uncovers how factors like temperature and solar radiation influence yields, providing a new tool for optimizing agricultural production.
The study introduces a varying-coefficient functional regression model (VCFRM) using FDA to investigate the influence of environmental factors like temperature and solar radiation on crop yields over time. The analysis of strawberry and tomato data revealed key periods when these factors most significantly impact growth.