TY - JOUR ID - 155484 TI - Combined Standardized Precipitation Index and ANFIS Approach for Predicting Rainfall in the Tropical Savanna Region JO - Journal of Soft Computing in Civil Engineering JA - SCCE LA - en SN - AU - Gandhi, Fenil R. AU - Patel, Jayantilal N. AD - Ph.D. Student, Department of Civil Engineering, S. V. National Institute of Technology, Surat, India AD - Professor (HAG), Department of Civil Engineering, S. V. National Institute of Technology, Surat, India Y1 - 2022 PY - 2022 VL - 6 IS - 3 SP - 63 EP - 77 KW - ANFIS KW - Drought KW - Forecasting KW - SPI KW - Surat DO - 10.22115/scce.2022.333365.1412 N2 - Climate change has affected many sectors in the world. Therefore, the Prediction of climatic factors is essential in case to achieve sustainability in human life. Rainfall prediction is also important as the agricultural sector depends on rainfall, and human life depends on agricultural products. This study presents the Standardized Precipitation Index (SPI) prediction using the adaptive neuro-fuzzy inference system (ANFIS). Various models (6 nos.) with different combinations of Rainfall and SPI values are prepared to predict the SPI index. Out of these six models, the M2 model (SPI3 SPI4 R4) performed best in the case of SPI 5. (RMSE value is 0.059, the R2 value is 0.987, and the value of the coefficient of determination is 0.993. In the case of SPI 6, the M1 model (SPI5 SPI4 SPI3 R5) performed best (RMSE value is 0.042, the R2 value is 0.992, and the value of the coefficient of determination is 0.996. The outcome may be helpful to the policymakers, scientists, researchers and government authorities in building a policy for sustainable water resources management in the region. UR - https://www.jsoftcivil.com/article_155484.html L1 - https://www.jsoftcivil.com/article_155484_6214c91e498d7e4128d8c5e254e18c3b.pdf ER -