This paper focuses on municipal solid waste generation in city of Tehran, the most populated city in Middle East. Three methods are explored in this paper to analyze the past solid waste time-series analysis: regression, Artificial Neural Network (ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS). | An intelligent algorithm for accurate forecasting of short term solid waste generation International Journal of Data and Network Science 1 2017 59 68 Contents lists available at GrowingScience International Journal of Data and Network Science homepage ijds An intelligent algorithm for accurate forecasting of short term solid waste generation Mohana Fathollahi Saeed Heidari Farsani and Ali Azadeh School of Industrial Engineering College of Engineering University of Tehran Tehran Iran CHRONICLE ABSTRACT Article history Municipal solid waste management has become a global concern during the past decades in Received October 1 2017 many countries such as Canada and waste management technological advancements and regu- Received in revised format No- lations have been increased. Solid wastes emit greenhouse gases which result in global climate vember 16 2017 change pollution of air and water which has tremendous negative impact on human health. Due Accepted May 21 2018 Available online to the excessive urbanization and fast economic development municipal solid wastes have been May 21 2018 increased in developing countries. In order to manage this emerging issue polluted countries Keywords need a series of legislations and policies toward solid wastes. Accurate prediction of future mu- Waste Prediction nicipal solid waste generation plays a critical role for future planning. This paper focuses on Municipal Solid Waste MSW municipal solid waste generation in city of Tehran the most populated city in Middle East. Three Regression approach methods are explored in this paper to analyze the past solid waste time-series analysis regres- Artificial Neural Network sion Artificial Neural Network ANN and Adaptive Neuro-Fuzzy Inference System ANFIS . Adaptive Neuro-Fuzzy Inference The first method which is the classical regression approach is used as a baseline for considered System neural networks models. The second method utilizes the past data as training
