Mathematical Modeling for Climate-Based Optimization of Rice Planting Schedules

The stability of rice production is greatly influenced by the dynamics of climate variability that changes rapidly and is unpredictable. This study developed a climate-based planting scheduling model that utilizes daily climate data and annual production data for the period 2016–2024. The predictive model was built through multiple linear regression to examine the effects of temperature, rainfall, humidity, and wind speed on crop yields and ARIMA to project climate and rice production until 2029. Data were obtained from BMKG, BPS, and related regional agencies, then processed to produce an adaptive planting schedule. The regression results showed high accuracy with R² = 0.99, Adjusted R² = 0.961, MAE = 5.980, and RMSE = 6.770. Rainfall showed a negative effect (p = 0.025) on rice production. The optimization model produced the two most profitable planting months each year and provided more stable yields than conventional planting patterns. Five-year production projections show fluctuations influenced by climate conditions, including a sharp decline in 2027 and a rebound in 2029. The development of an adaptive schedule model allows for alternative decision-making in areas vulnerable to climate change.