Author(s): Nikolaos Mellios; Dimitris Kofinas; Elpiniki Papageorgiou; Chrysi Laspidou
Keywords: Water Demand; ANFIS; Skiathos; Water Distribution Network; Multivariate Analysis
Abstract: Considering the increasing demand for the optimization of water distribution networks in terms of leakage reduction and pressure management, as well as the need to reduce urban water consumption, a lot of effort has been invested in the past decade in order to define accurate, long-term and short-term water demand forecasting methods. Linear regression models, such as ARIMA, and Artificial Neural Networks (ANN) have been used, as well as different hybrid approaches. In this paper, a multivariate analysis of daily water demand of Skiathos Island, Greece and an investigation on the benefits of the Artificial Neuro-Fuzzy Inference System (ANFIS) forecasting method are presented. Skiathos is a touristic island in the Aegean Sea with typical Mediterranean climate and seasonally intense population fluctuations due to touristic activity. These parameters form a highly periodic water demand, with summer demand surpassing by far—almost six times—winter demand. The applied methodology considers how touristic activity and meteorological and hydraulic variables influence water demand. The applied method benefits when water demand includes non-linear parts and performs adequately; it also provides a Fuzzy Rule Base, giving researchers and water managers a handy tool for interpreting the physical aspects of the inter-relationships.