Estimation of Domestic Solid Waste Amount with Exponential Smoothing Method and Artificial Neural Network Models: An Application for Istanbul Province

Akgül M., Kalaycı E. E. , Doğan B. , Çalış Uslu B.

Innovations in Intelligent Systems and Applications Conference (ASYU), 15 - 17 Ekim 2020, ss.1-6


Population growth, urbanization and industrialization in the globalizing world also brings with it the waste problem Wastes that cannot be properly stored, collected and disposed pose an important threat to public health and the environment. In municipal solid waste (MSW) management, estimating trends and assessing their impacts play a key role in planning and implementing ecologically sustainable strategies.Within the scope of this study, based on the district-based domestic solid waste data of Istanbul Metropolitan Municipality between 2004-2019, the estimation of the domestic waste amounts for 2020 was made by exponential smoothing (ES) and artificial neural networks (ANN). The performance of ANN and ES models was evaluated using the mean absolute percent error (MAPE). The accuracy of the models was tested with a case study in 39 districts in Istanbul Metropolitan. Results showed that the ANN, as a non-linear model, has a higher predictive accuracy than exponential smoothing model.