The Research On Investment Cost Estimation Of Y Construction Project

The construction industry is one of the most rapidly developing industries in recent decades,and the economic benefits it generates are still a major part of China’s national economy.However,with the slowdown of the national economic development trend and the deepening of the transformation and reform of the construction industry,the construction industry has entered a phase of flat development and industry consolidation,and many projects are now facing problems of uncontrolled investment,financing difficulties and poor quality.According to the market research,it is known that the main reasons leading to wasteful and out-of-control investment include the wrong decision in the pre-construction period,unreasonable cost control in the construction period,etc.In the whole life cycle of the construction project,the investment decision stage has the highest degree of influence on the construction project investment cost control,and the investment estimate is an important basis for the pre-investment decision and the later cost control,which can provide data for the project feasibility study and design plan comparison reference.Therefore,this paper selects Y construction project,which is currently in the investment decision stage,as the research object for the study of investment cost estimation.In this paper,the first 26 influencing factors of Y construction project investment cost are identified by expert survey method,and the top 20 influencing factors with higher weight and influence are calculated by using hierarchical analysis method as the input vectors of Y construction project investment cost estimation model(i.e.BP neural network model).Then,the BP neural network model is established,and the sample data of 21 construction projects of the same type are substituted into the model for training,and the algorithm and parameters of the BP neural network model are continuously adjusted according to the training effect to obtain the final improved BP neural network model,and four groups of samples were randomly selected for estimation comparison,and the relative error between the predicted and actual values of the samples did not exceed 2%at most,which was much lower than the error requirement within ±10%of the investment estimation at the feasibility study stage.Finally,the actual project data of Y construction project is substituted into the improved BP neural network model,and the estimated value of investment cost of Y construction project is obtained quickly and accurately,and by adjusting the parameters of each influencing factor,different investment costs are obtained for comparison and analysis,and the investment cost proposal of Y construction project is given.In summary,the improved BP neural network-based investment estimation model built in this paper relies on MATLAB toolbox,which is fast and simple to operate,improves the accuracy and computing speed of investment cost estimation of Y construction project and also improves the quality of investment decision,provides an important basis for project financing scheme,fund raising and subsequent construction drawing design and cost management,and enhances the enterprise Project management level and economic efficiency.