ANFIS for rice yelds forecasting

Abstract

Almost 90% of rice is produced and consumed in Asia, and 96% in developing countries. In Malaysia, the Third Agriculture Policy (1998-2010) was established to meet at least 70% of Malaysia’s demand a 5% increase over the targeted 65%. The remaining 30% comes from imported rice mainly from Thailand, Vietnam and China (Saad et al., 2006). Raising level of national rice self-sufficiency has become a strategic issue in the agricultural ministry of Malaysia. The ability to forecast the future enables the farm managers to take the most appropriate decision in anticipation of that future. The accuracy of time series forecasting is fundamental to many decisions processes (Zou et al., 2007). One of the most important and widely used time series model is artificial neural network (ANN). ANN is being used more frequently in the analysis of time series forecasting, pattern classification and pattern recognition capabilities (Ho et al., 2007). ANN provides an attractive alternative tool for both forecasting researchers and has shown their nonlinear modeling capability in data time series forecasting. Another approach is using fuzzy logic. Fuzzy Logic first developed to explain the human thinking and decision system by Zadeh (Sen et al., 2006). Several studies have been carried out using fuzzy logic in hydrology and water resources planning (Chang et al., 2001; Liong et al., 2000; Mahabir et al., 2000; Ozelkan and Duckstein, 2001; Sen et al., 2006). Recently, an adaptive neuro-fuzzy inference system (ANFIS), which consists of the ANN and fuzzy logic methods, has been used for several application, such as database management, system design and planning/ forecasting of water resources (Chang et al., 2006; Chang et al., 2001; Chen et al., 2006; Firat and G¨ung¨or , 2007; Firat, 2007; Nayak et al., 2008). The main purpose of this study is to investigate the applicability and capability of the ANFIS and ANN for modeling of rice yields time-series forecasting. To verify the application of this approach, the rice yields data from 27 stations in Peninsular Malaysia is chosen as the case study.