This study predicted maximum rice yield in Davao de Oro using time series models: autoregressive integrated moving average (ARIMA), radial basis function neural network (RBFNN), and adaptive neuro-fuzzy inference system (ANFIS). Quarterly rice production data from 2004 to 2024, sourced from DA-PHILRICE, were analyzed to determine production trends, develop best-fit models, and compare forecast accuracy. RBFNN and ANFIS closely follow the actual test data values, demonstrating their ability to capture fluctuations in rice produce. Based on the result, RBFNN is the most accurate model. These findings suggest a need for the Department of Agriculture to implement strategic interventions, including precision agriculture, enhanced irrigation, distribution of climate-resilient varieties, and farmer training on adaptive techniques and resource efficiency, to mitigate the projected decline of rice yields. Action plan and policies were recommended to improve the rice production with series of activities in the province.

Keywords: rice produce, autoregressive integrated moving average (ARIMA), radial basis function neural network (RBFNN), adaptive neuro-fuzzy inference system (ANFIS), Davao de Oro