Abstract
The growth of lettuce (Lactuca sativa) in controlled environments such as Plant Factories is highly influenced by lighting, particularly under red-blue (RB) LED treatment. Accurate growth prediction models are essential for optimizing yield. This study compared four models linear, polynomial, logistic, and Gompertz to determine the best predictor of leaf area expansion. Leaf area measurements over 30 days were analyzed using Easy Leaf Area software. Results showed that the Gompertz model consistently outperformed others with the lowest Mean Absolute Percentage Error (MAPE) of 14.55% (slow), 39.51% (medium), and 29.13% (high), and the highest R² values of 0.99 across all growth categories. In contrast, linear and polynomial models exhibited extremely high MAPE values, exceeding 300% in most cases. The study concludes that the Gompertz model is the most accurate and biologically realistic for modeling lettuce growth in Plant Factory systems, offering robust predictive capability for sustainable precision agriculture.