Using RBF Neural Network in Forecasting Urban Construction and Demolition Waste Generation

To achieve the effective construction and demolition waste (C&DW) management, it is of great significance to predict CD&W generation accurately to carry out the recycling treatment and formulate relevant policies for authorities. On the basis of the univariate time series of limited data samples, this paper puts forward a radial basis function neural network (RBFNN) to solve the C&DW forecasting problem, which involves the algorithm of network training and prediction process. Three indicators are used to comprehensively evaluate the performance of the proposed method applicated in Shanghai and Hong Kong cases. Compared with other forecasting techniques, such as multiple linear regression (MLR), support vector regression (SVR), and back-propagation neural network (BPNN), this paper verifies that the RBFNN prediction model has strong accuracy in the analysis of C&DW generation prediction.