基于自适应模糊神经网络的建筑工程造价预测研究-管理科学与工程专业论文.docxVIP

Abstract The whole life cycle ∞st management is a trend in the future，the key to the dynamic project cost management is to control each stage cost of construction project ， which contains inves位nent estimate，design estimate，construction drawing budget and settlement，to prevent design estimates exceed investment estimates ，budget exceed design estimates and final accounts exceed budget. Therefore，the construction cost prediction is important in the feedforward managemen t. This dissertation aims to study a fast，accurate，convenient，and authentic model to predict project cost，which can provide the basis for the whole life cycle cost management. Currently，there are some problems in Chinas project cost prediction. In terms of technology，first，the existing models fail to express the nonlinear relationship between construction characteristics and project ∞st，the result is lack of persuasion; second，the existing intelligent control theory and method in the field of project ∞st prediction is less. in t臼ms ofmanag田nent，one is in the 町a ofbig dat龟 construction ∞st data collection system is not perfect ，construction classification standards have not been fully extended，resulting the massive construction cost data is difficult to data mining and analysis; second is construction project involving a number of professional projects ，and the project ∞st hωmany influence factors，a construction project cost index system is required. Therefore，through a large number ofliterature review，and study the project cost management and forecasting methods 企om home and abroad，this dissertation is on the results of previous studies，in ord町 to sort out a set of complete classification standards for construction projects. Build an index system of construction cost. Build a building project ∞st prediction model by ANFIS. Through 臼npirical study，the model is proved to have the advantages of convenient operation ，fast and accurate results. It shows that 由is method is fe