Abstract
Taiwan’s agriculture is facing global competition and challenge due to Taiwan’s entry to World Trade Organization (WTO). Being a WTO member, Taiwan needs to open its agriculture market by removing the tax barrier. Agriculture production is more complex than industrial production, as it is influenced by many unexpected factors, such as rain quantity, rain acidity, sunshine, and temperature. The development and application of an accurate demand-supply forecasting model is one way to increase Taiwan agriculture’s competitiveness. This research uses Artificial Neural Network (ANN) to develop a forecasting model to predict the supply of large-scale vegetable and fruit. The effects of twelve factors related to climate and economy on the production quantity per acre are analyzed. Based on the historical data, the proposed ANN model shows its capability to accurately forecast the production quantity per acre for both vegetable and fruit. The use of this model can help farmers to forecast their production quantity to better react to the climate changes and to match the demand requirement from the market.