Prediction of Municipal Solid Waste Generation by Use of Combination of Artificial Neural Network and Principal Component Analysis

摘要

Accurate prediction of quantity of municipal solid waste (MSW) is crucial for designing andprogramming MSW management systems. But predicting the amount of generated waste is adifficult task because various parameters affect it and also its fluctuation is high. In this study,an appropriate model was proposed using feed forward artificial neural network (ANN) forpredicting the weight of waste generation in Mashhad. In addition, since there were manyvariables in this research as inputs to ANN model, principal component analysis (PCA)technique was used to reduce the number of inputs for ANN model (PCA-ANN). The PCAwas reduced the input variables from 13 to 8. Then the ANN model with 8 input variableswas developed. Finally, ANN and PCA-ANN models were compared with each other whichresulted in improving through preprocessing on input variables. Furthermore, PCA-ANN wasthe superior model because it possessed a faster training speed and a more satisfactorypredicting performance.