| With the rapid development in recent years, China’s construction industry has gradually become a pillar industry to promote the development of the economy. But compared with the rapid development of our construction industry, the engineering quality management research is relatively lagging behind. Project quality is the life of the building products.The quality of the engineering not only affects the applicability of the engineering, but also people’s life and property safety, and even social harmony and stability.Most of the engineering quality control method in our country are in-process control or afterwards control, while advance quality control method is used less.This article set about from analyzing the five major factors affecting the quality of engineering, after selection of quality control points, establishing the index system of engineering quality control. By introducing neural network, the engineering quality control model based on neural network is established. Using the strong self-learning ability of the neural network on the collected engineering training samples, it is concluded the weight of quality control points according to the results of the training sequence, and then put forward some reasonable suggestions for the key influencing factors.Neural network used in engineering quality control can be very good to make up for the deficiencies of traditional quality control methods.It is able to solve complex nonlinear analysis of the problem of large quantities of data.It avoids the shortcomings of subjective determination in determining critical control points.By training the neural network functions and relevant parameters improved to make it more efficient and accurate training data and get the right value of the result, it is possible to .provide precise control of the basis for quality control of construction enterprises.During the construction process, we need to focus on controlling the level of experience in technical management personnel, the factors of construction enterprise quality management system and so on, while other aspects of the selected schedule control and testing organizations qualified we can point with less effort relatively. |