Analysis and Forecast of Construction Waste Based on ARIMA Model

Abstract

In the construction waste output forecast, it is difficult for forecasting and reducing the accuracy of forecasting. In this paper, a method for estimating the yield of urban construction waste is built on the basis of building area, by using the ARIMA model to predict demolition area and the area of the building was completed, from the empirical coefficient method and indirect method to predict for construction waste production. Haikou as an example to calculate and predict the 2015-2020 years of urban construction waste output. Overvie w of related forecasting methods for construction waste In recent years, with the accelerated pace of urban construction, construction waste will be generated with a large number of production. According to the source of construction waste and the way of producing, construction waste can be divided into three categories, in the construction site of construction waste, public buildings and residential decorative decoration construction waste and housing demolition (1) . The muck from construction sites can basically reach the balance of supply and demand. The processed resources mainly for decoration waste, demolition waste and garbage remaining under construction. At present, the prediction model of construction waste generated mainly includes Multiple Regression Model, Grey Model and Time Series Prediction. Multiple Regression Model is used to combine the urban population, urban residents' income, and then to predict the future construction waste output (2) . Grey system processing is using a certain mathematical method to eliminate the influence which based on the analysis of the observed data the establishment of the system can make a prediction of the Gray Model (3) . Through the process of forecasting target's own time series to study the forecast target itself on the change trends, so the prediction of the future of the data is a Time Series Model Prediction method (4) . These existing estimation method, however, there are factors to consider not to comprehensive, the data quantity demand is high or relatively poor commonality problem and so forth. Therefore, it is necessary to build a set of general method for the estimation of urban construction waste output based on the existing conditions. The ARIMA model based on time series analysis method in this paper, introducing the difference parameters , in order to solve the data loss caused by the imbalance in non-stationary time series.