Study On The Prediction Of Construction Cost Based On Uncertain Linear Regression

In market economy system, with the development of construction project pricing method, prediction of construction cost has become an important part of project cost management. On the one hand, as the Project cost prediction is affected by the characteristics of the project cost, so to determine the object of study and identify the specific effects of strong related factors of project cost has become the primary problem. On the other hand, due to the lack of historical data, neural network, fuzzy set, grey forecasting and time series method can not ensure the precision of prediction. Therefore, the determination of strong related factors of project cost and the improvement of prediction accuracy is a significant thing. The research of project cost prediction is the need of theory and Practice.This paper qualitatively analyzes the macro impact factors of construction project cost in the design stage,and selects the shear wall structure of high-rise residential building of Handan as the study object. According to the constitution of Hebei’s construction and installation costs and the calculation procedure of project cost. Based on stepwise regression method, the main micro factors are subject to multiple correlation analysis. we obtain the strong micro factor of project cost by multiple correlation analysis of main micro factors. Then we introduce the regression analysis theory into uncertainty theory, use Delphi method to analysis the experts’ experience data, and build uncertain linear regression model. At the end of the paper, a example of Handan high-rise residential building(civil part) is provided to illustrate the effectiveness of the model.The application of the prediction model offer help for the construction, the developer, the Contractor and the building works of any third-party.