Water Requirement Prediction System using Multi-Factors High Order Fuzzy Time Series Method

Madura Island is a dry area. Madura Island has a deficient average rainfall of 10.5 mm per year. Bangkalan is one of the areas on the island of Madura with a high population density compared to other Madura areas. This causes a shortage of water for daily consumption. Thus, forecasting is needed to predict the water requirement in Bangkalan for clean water service companies. The method used is Multi-Factors High Order Fuzzy Time Series (MFHO FTS), where this method is the Fuzzy Times Series (FTS) method which employs more than one factor that affects forecasting. This study’s contribution is the water requirement prediction system using MFHO FTS Method. There are several steps in the Multi-Factors High Order Fuzzy Time Series method, including forming a Universe of Discourse, determining the number of clusters, forming subintervals, forming fuzzy sets, the fuzzification processes, forming Fuzzy Logic Relationships (FLR), and the defuzzification processes. The measurement with the MAPE calculation obtained from this study is 0.09. The MAPE value close to zero indicates that the Multi-Factors High Order Fuzzy Time Series method can make predictions well.