摘要
为了准确预测水运工程人工单价,使定额人工单价与市场人工单价更好接轨,构建基于BP神经网络的水运工程人工单价动态预测模型,并从隐含层数、隐含节点数两个方面对模型进行优化。最后以优化模型对2020年海南省水运工程综合人工单价进行预测作为验证,预测结果为137.76~142.68元/工日。结果表明:该网络模型具有较好的泛化能力,不仅可以弥补各种方法的不足,提高预测方法的适用性和实用性,而且可以及时有效的对定额人工单价进行更新,能够准确地对人工工日单价予以合理准确的计算,并具有一定的推广意义。
In order to accurately predict the labor unit price of water transportation projects and make the fixed labor unit price better integrate with the market labor unit prices,a dynamic prediction model of the labor unit price of water transportation projects based on BP neural network is constructed.The model is optimized from two aspects:the number of hidden layers and the number of hidden nodes.The model is used to predict the comprehensive labor unit price of Hainan Province’s water transportation project in 2020 as the verification,and the prediction result is 137.76~142.68 yuan/man-day.The results show that the network model has a good generalization ability,not only can make up for the shortcomings of various methods and improve the applicability and practicability of the forecasting method,but also can timely and effectively update the fixed labor unit price.The unit price per working day is calculated reasonably and accurately,and it has certain promotion significance.
作者
汪优
马婷婷
王天明
陈宝光
WANG You;MA Ting-ting;WANG Tian-ming;CHEN Bao-guang(College of Civil Engineering,Central South University,Changsha 410075,China;Beijing Zhongjiao Jingwei Engineering Technology Co.Ltd.,P.R.C,Beijing 100025,China)
出处
《工程管理学报》
2021年第2期39-43,共5页
Journal of Engineering Management
基金
海南省交通运输厅项目(SZEWEC-2019-1015).
关键词
水运工程
人工单价
BP神经网络
预测
SPSS
water transportation project
labor unit price
BP neural network
prediction
SPSS
