Forecasting municipal solid waste generation using artificial intelligence models—a case study in India

Abstract

Proper management of municipal solid waste is one of the prime matters of concern for metropolitan cities. To be able to successfully manage the solid waste generated, we need to plan in advance. A very essential pre-requirement for an efficient solid waste management is an accurate prediction of the garbage generation. Accurate forecasting of the quantity of MSW generation will enable us to design and operate an effective waste collection system. The main objective of this paper is to compare different models of artificial intelligence, viz. artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), discrete wavelet theory–artificial neural network (DWT-ANN), discrete wavelet theory–adaptive neuro-fuzzy inference system (DWT-ANFIS), genetic algorithm–artificial neural network (GA-ANN) and genetic algorithm–adaptive neuro-fuzzy inference system (GA-ANFIS) to examine and evaluate their capability in forecasting the amount of garbage being generated. A case example of the city of New Delhi, India, has been used for better understanding of different models. Root mean square error (RMSE), coefficient of determination (R2) and index of agreement (IA) values for every model were calculated, and the models were compared on the basis of it. The hybrid model of genetic algorithm and artificial neural network was found to have the lowest RMSE, the highest IA value and the highest R2 values, and hence is the most accurate of the above six models.