Construction Waste Production Prediction and Correlation Analysis Based on ARIMA-BP Combination Model

Abstract

Reliable construction waste production data are prerequisites for construction waste management and resource utilization. However, since there is no relevant construction waste production data in various cities in China, it is necessary to calculate and predict construction waste production in different cities. This paper aims to verify a reliable prediction method of construction waste production based on the data in the statistical yearbook. Through the study, it is found that ARIMA-BP combination forecasting model has shown ideal accuracy in the prediction of construction waste production in Tianjin, Kunming and Xining, and the average relative error has reached about 0.01. This paper shows that in the next few years, the production of construction waste in the three cities has increased to varying degrees, and the highest production of construction waste in Tianjin in the next few years will reach more than 35 million tons. At the same time, by analyzing the composition proportion and influencing factors of construction waste, it is hoped that this paper will provide help for the management and resource utilization of construction waste.