Research On Forecasting System For Construction Cost Based On J2EE

With the development of construction industry, the management of construction projects becomes more important. The forecasting of construction cost is the basis of the decision-making and construction cost control for construction projects, its accuracy often erects the most influences on the investment amounts and progress of construction projects. It is sorely urgent to develop an efficient and applicable method of forecasting for construction cost in the construction industry.This paper first introduces several common construction cost forecasting methods, such as BP neural network and grey system theory, analyzes their advantages and disadvantages. A new method based on K-Modes clustering method is put forward for its benefit on dealing with missing data and text data in the real application. The results show that the new method can meet practical demand.After the division of the modules in forecasting system, a web report generation method based on POI is introduced. We choose Struts and Hibernate as the architecture, design and realize a construction cost prediction system based on J2EE.