政府投资项目的BPNN、Delphi、GM(1,1)投资估算方法  

Investment Estimation of Government Projects on the Basis of BPNK,Delphi and GM(1,1)

作　　者：段晓晨[1] 邢红宏[2] 张小平[1] 

机构地区：[1]石家庄铁道学院经管分院,河北石家庄050043 [2]海军航空工程学院基础部,山东烟台261001

出　　处：《工业工程》2007年第5期85-88,101,共5页Industrial Engineering Journal

基　　金：国家自然科学基金资助项目(70373032)

摘　　要：运用Delphi法,对专家经验进行分析整理预测无类似项目的拟建工程分部分项工程造价,GM(1,1)法预测具有小量类似项目数据的分部分项工程造价,同时充分利用了BP神经网络的"提取器"作用,在大量具有类似已完工程数据资料中"提取"和"挖掘",实现了对具有大量类似项目的分部分项工程造价的准确估算。通过案例实证说明,预测值与实际值的相对误差很小,足以满足造价估算的要求。This paper applyies Delphi method to estimate the scheduled items without similar projects based on the experts＇experience,and GM（ 1,1 ） to forcast the items＇costs with few similar projects. With plethora of the historic data, artificial neural network based on BP algorithm provides a practical solution for that problem and it is applied to determine cost estimation through ＂distilling＂ rules from the materials of completed projects, From the example, it can be seen that the relative errors are small enough for accuracy demand of cost estimation after simulation, and test result shows the model based on BP neural network is accurate and successful.

关 键 词：造价估算 BP神经网络 投资控制 DELPHI法 GM(1 1)法 

分 类 号：F830[经济管理—金融学]