Research On Civil Engineering Cost Estimation Method Of Subway Based On BPNN

Following our country social economy development and urbanized advancement speeding up, the subway engineering has already become an important development domain in our country city infrastructure construction. But, due to lacking engineering construction fund in our country and its soaring cost, the development of subway construction has been actually limited. Therefore, how to take effective measures to reduce the subway engineering cost has been already become an important and chief research topic that can speed up the construction of subway engineering.The early stage of the project, in which the construction cost estimation of feasibility study is the basis of the decision-making and construction cost control for the project, often erects the most influences on the cost management of construction project. So it is vital to find a feasible and efficient method that could be used for cost estimation of construction projects. In the recent years, the artificial neural networks carries on the simulation to the basic characteristic of the human brain or the natural nerve network, has the very extensive application in the field of engineering cost estimation.Based on great amount of investigation and deep analyze of cost estimation methods and models of construction project both domestic and abroad, in the thesis a model has been set up, making use of the coalescence of the fuzzy math and BP neural network, according to the characters of the subway engineering cost. In order to enhance the accuracy of the project cost estimating model, samples of subway engineering cost estimating model are selected using fuzzy math method. Regarding to civil engineering cost estimation model of subway based on BP neural network has been built, the model based on BPNN has been trained, simulated and examined under the circumstance of MATLAB language. In the end, the model which has been set up in this thesis is tested by the data of the factual project. The result indicates the BP neural network model is of better capability of generation, it can accurately estimate the civil engineering cost of subway.