Research On Intelligent Calculation Of Construction Cost Index

In July 2020,the circular of the General Office of the Ministry of Housing and urban-rural development on the issuance of the programme of work for the reform of construction costs stipulated the abolition of the quota-based pricing of maximum bid prices and the phasing out of the issuance of budget quotas,emphasizing market-based pricing,the engineering cost index plays an increasingly important role in the engineering construction management.At present,the calculation method of construction cost index is heavy in work and slow in speed.It is of great practical value to study the fast and accurate calculation method of construction cost index.This paper expounds the theory of project cost and the concept of project cost index,determines the calculation set of project cost index considering the need of project practice,and analyzes the applicability,advantages and disadvantages of the existing calculation methods of project cost index,the GA-BP neural network is selected as the method to calculate the cost index of building engineering.Based on the systematic analysis of the engineering characteristics which affect the construction cost index,14 engineering characteristics which describe the construction project mode are selected from three aspects of the regional index,the time index and the technical index,according to the applicability of the selected feature indexes in different building types and different phases of the project,the engineering feature set of different building types in different phases is established.In view of the selected project characteristic index,through the detailed project cost analysis,obtains the project characteristic change to the project cost index influence rule,more reasonably determines the project characteristic quantification method,it realizes the quantification of different kinds of engineering features.According to the determined construction project characteristics and project cost index set,80 completed residential project characteristics and cost index related data are collected and collated,the construction characteristic index database and the cost index database of the completed project are established.Based on GA-BP neural network,this paper establishes the estimation model of construction cost index through training.Ten six-storey and 11-storey residential projects are selected as study samples,and18-storey and 27-storey residential projects are selected as test samples,the rationality of this quantitative method is verified by comparing the error of one-sided cost estimation,and taking 10 housing projects as test samples,the project cost index is calculated,and the result is compared with the actual data,the practicability of the model is verified.The sub-estimation model is intelligent and can realize the fast prediction of project cost index.The calculation method of the project cost index established in this paper is fast and accurate,and has the value of engineering practice.