国际工程投标报价中标高金预测的高斯过程方法应用
(中国港湾工程有限责任公司, 北京 100027)
- 文献要素
-
[摘 要] 针对国际建筑工程项目投标报价中最优标高金的确定问题,提出了一种基于高斯过程方法的标高金预测模型。在分析标高金影响因素的基础上,采用高斯过程方法对标高金影响因素和最优标高金数值之间的复杂非线性关系进行建模,并对30个典型国际建筑工程案例进行实证研究。实验结果表明,与现有的人工神经网络、支持向量机等方法相比,该方法具有较高的预测精度和较低的时间复杂度。[关键词] 国际工程项目;投标报价;标高金预测;高斯过程Abstract: A novel Bid Mark-up of International Construction Projects Estimation model is proposed, which is based on the Gaussian Processes algorithm. The various factors that may affect the Mark-up estimation in a project have been identified, and the complex nonlinear relationship between the factors and the values of mark-up is modeled using Gaussian Processes. Moreover, the mark-ups of 30 typical international projects are conducted by a real case study. The experimental results show that the presented method has high accuracy and low time complexity compared with traditional methods, ANN and SVM et al.Key words: international project; bidding; mark-up estimation; Gaussian processes[参考文献][1]Heng L. Neural network models for intelligent support of markup estimation[J]. Engineering, Construction and Architectural Management,1996(3):69-81.[2]王雪青,喻刚,孟海涛.基于GA改进BP神经网络的建设工程投标报价研究[J].土木工程学报,2007,40(7):93-98.[3]喻刚,王雪青,赵辉.基于RS与ANFIS的投标报价决策研究[J].湖南大学学报:自然科学版,2008,35(5):89-92.[4]任玉珑,唐道鸿.投标报价中报高率确定的支持向量机方法研究[J].科技管理研究,2006(11):237-241.[5]Rasmussen C. E.Gaussian processes for machine learning[M].Massachusetts: The MIT Press,2006:35-50.[6]王雪青.国际工程投标报价决策系统研究[D].天津:天津大学,2003.建筑经济,2013(6):28-30