«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《建筑科学与工程学报》[ISSN:1673-2049/CN:61-1442/TU]

卷:

40卷

期数:

2023年04期

页码:

125-134

栏目:

桥梁工程

出版日期:

2023-07-10

文章信息/Info

Title:

Cost model prediction of minor repair project of bridge and culvert based on machine learning method

文章编号:

1673-2049(2023)04-0125-10

作者:

史小丽¹,王晓伟²,邱晨阳³,高 楠⁴

(1. 长安大学 公路学院,陕西 西安 710064; 2. 中国路桥工程有限责任公司,北京 100011; 3. 苏交科集团股份有限公司,江苏 南京 210019; 4. 陕西交通控股集团有限公司,陕西 西安 710065)

Author(s):

SHI Xiaoli¹, WANG Xiaowei², QIU Chenyang³, GAO Nan⁴

(1. School of Highway, Chang'an University, Xi'an 710064, Shaanxi, China; 2. China Road & Bridge Corporation, Beijing 100011, China; 3. JSTI Group Co., Ltd., Nanjing 210019, Jiangsu, China; 4. Shaanxi Transportation Holding Group Co., Ltd., Xi'an 710065, Shaanxi, China)

关键词:

桥涵资产;  小修费用;  灰色关联度;  岭回归;  Lasso回归

Keywords:

bridge and culvert asset;  minor repair cost;  grey correlation degree;  Ridge regression;  Lasso regression

分类号:

U445.2

DOI:

10.19815/j.jace.2021.12106

文献标志码:

A

摘要:

基于陕西省11条高速公路2008～2015年桥涵小修工程量清单历史数据,采用机器学习算法,以组成桥涵各构件的小修费用作为因变量,研究桥涵小修费用预测模型。通过灰色关联度模型分析桥涵小修费用的影响因素,采用皮尔逊相关系数检验法对各影响因素进行多重共线性检验,筛选出解释变量,使用岭回归和Lasso回归对桥涵各构件小修费用模型进行回归分析,得到桥涵小修总费用预测模型。结果表明:桥涵小修工程费用的影响因素主要有通车年限、桥梁(涵洞)长度、年平均当量轴次、桥涵所处地区的年均降雨量和温度及车道数; 基于模型预测所依托高速公路2016～2017年桥涵小修费用,并与该年度实际费用进行Wilcoxon符号秩检验,检验结果均大于0.05,验证了机器学习法预测桥涵小修工程费用模型的有效性,预测结果能为分配养护费用、提高养护决策水平提供合理建议。

Abstract:

Based on the historical data of bill of quantities of bridge and culvert minor repair project on 11 expressways in Shaanxi province from 2008 to 2015, the machine learning algorithm was used to study the prediction model of bridge and culvert minor repair cost with the minor repair cost of each component of bridge and culvert as the dependent variable. Through the grey correlation degree model, the influencing factors of the minor repair cost of bridge and culvert were analyzed. The Pearson correlation coefficient test method was used to test the multicollinearity of each influencing factor, and the explanatory variables were selected. Ridge regression and Lasso regression were used to analyze the minor repair cost model of each component of bridge and culvert, and the total cost prediction model of minor repair of bridge and culvert was obtained. The results show that the main influencing factors of the cost of bridge and culvert minor repair projects are years of operation, length of bridge(culvert), annual average equivalent axles, annual average rainfall and temperature in the area where the bridges and culverts are located, and number of lanes. Based on the model prediction, the bridge and culvert minor repair cost of the expressway from 2016 to 2017 is carried out, and the Wilcoxon signed rank test is carried out with the actual cost of the year. The test results are all greater than 0.05, which verifies the effectiveness of the machine learning method to predict the bridge and culvert minor repair project cost model. The prediction results can provide reasonable suggestions for allocating maintenance costs and improving maintenance decision-making level.

参考文献/References:

[1] 陈树礼,刘永前.城市与公路桥梁技术状况评定方法对比分析[J].建筑科学与工程学报,2018,35(3):95-103.
CHEN Shuli,LIU Yongqian.Comparative analysis of technical condition evaluation of city bridges and highway bridges[J].Journal of Architecture and Civil Engineering,2018,35(3):95-103.
[2]FRANGOPOL D M.Life-cycle performance of deteriorating structures:assessment,design,and management[M].Reston:American Society of Civil Engineers,2003.
[3]牛宏亮.城市桥梁养护费用预测模型与方法研究[D].广州:华南理工大学,2009.
NIU Hongliang.Research on prediction model and method of urban bridge maintenance cost[D].Guangzhou:South China University of Technology,2009.
[4]ALI GHAHARI S,VOLOVSKI M,ALQADHI S,et al.Estimation of annual repair expenditure for interstate highway bridges[J].Infrastructure Asset Management,2019,6(1):40-47.
[5]JEBREEN J M I,JOHNSTON D W.Bridge maintenance level of service optimization based on an economic analysis approach[R].Raleigh:North Carolina State University,1995.
[6]MOLONEY M,MCKENNA T,FITZGIBBON K,et al.Quality data for strategic infrastructure decisions in Ireland[J].Infrastructure Asset Management,2017,4(2):40-49.
[7]张春霞,王 巍.基于费用效益最优的桥梁预防性养护时机确定方法研究[J].公路,2013,58(7):259-262.
ZHANG Chunxia,WANG Wei.Study on the method of determining the opportunity of bridge preventive maintenance based on cost-benefit optimization[J].Highway,2013,58(7):259-262.
[8]WOLDEMARIAM W,MURILLO-HOYOS J,LABI S.Estimating annual maintenance expenditures for infrastructure:artificial neural network approach[J].Journal of Infrastructure Systems,2016,22(2):04015025.
[9]戚云生.高速公路日常养护工程预算定额及费用标准研究[D].西安:长安大学,2010.
QI Yunsheng.Study on the budget quota and expenses standard of highway daily maintenance and engineering[D].Xi'an:Chang'an University,2010.
[10]乔建刚,侯会学.基于全寿命周期成本的桥梁改造方案优选的研究[J].中外公路,2014,34(5):93-95.
QIAO Jiangang,HOU Huixue.Research on optimization of bridge reconstruction scheme based on life cycle cost[J].Journal of China & Foreign Highway,2014,34(5):93-95.
[11]吴宇蒙,常 军.基于养护管理系统的既有桥梁生命周期成本模型研究[J].苏州科技学院学报(工程技术版),2010,23(1):44-49,55.
WU Yumeng,CHANG Jun.Study on in-service bridge life cycle cost model based on maintenance management system[J].Journal of Suzhou University of Science and Technology(Engineering and Technology edition),2010,23(1):44-49,55.
[12]朱 岩.陕西省高速公路桥涵小修保养年公里生产消耗指标研究[D].西安:长安大学,2019.
ZHU Yan.Research on the annual production and consumption of the maintenance project of the highway bridge-culvert in Shaanxi province[D].Xi'an:Chang'an University,2019.
[13]董金爽,隋 龑,薛建阳.传统风格建筑混凝土梁-柱节点承载力影响参数灰色关联分析[J].建筑科学与工程学报,2019,36(6):80-87.
DONG Jinshuang,SUI Yan,XUE Jianyang.Grey relevance analysis of influencing parameters of bearing capacity of concrete beam-column joints in traditional style buildings[J].Journal of Architecture and Civil Engineering,2019,36(6):80-87.
[14]孙作安,王 奇,张铁岩,等.分辨系数的作用及其确定方法[J].辽宁工学院学报,1997,17(4):71-72,84.
SUN Zuoan,WANG Qi,ZHANG Tieyan,et al.Function by resolution factor and its determination method[J].Journal of Liaoning Institute of Technology,1997,17(4):71-72,84.
[15]汪冬华.多元统计分析与SPSS应用[M].上海:华东理工大学出版社,2010.
WANG Donghua.Multivariate statistical analysis and SPSS application[M].Shanghai:East China University of Science and Technology Press,2010.
[16]陕西省统计局,国家统计局陕西调查总队.陕西统计年鉴2012[M].北京:中国统计出版社,2012.
Shaanxi Provincial Bureau of Statistics,Survey Office of the National Bureau of Statistics in Shaanxi.Shanxi statistical yearbook 2012[M].Beijing:China Statistics Press,2012.
[17]伍德里奇.计量经济学导论:现代观点[M].3版.北京:清华大学出版社,2007.
WOOLDRIDGE J M.Introductory econometrics:a modern approach[M]3rd ed.Beijing:Tsinghua University Press,2007.
[18]公路沥青路面设计规范:JTG D50—2017[S].北京:人民交通出版社,2017.
Specifications for design of highway asphalt pavement:JTG D50—2017[S].Beijing:China Communications Press,2017.
[19]陈永秀.相关系数含义的理解[J].中国考试,2011(7):15-19.
CHEN Yongxiu.Methods for calculating the correlation coefficient[J].Journal of China Examinations,2011(7):15-19.
[20]姚弘智,钟宝荣.基于机器学习线性回归的正规方程分析[J].电脑迷,2018(3):208.
YAO Hongzhi,ZHONG Baorong.Normal equation analysis based on machine learning linear regression[J].PC Fan,2018(3):208.
[21]刘 哲,乌 伟,张善文,等.基于多元岭回归估计小麦种植密度[J].中国农业科技导报,2020,22(6):72-80.
LIU Zhe,WU Wei,ZHANG Shanwen,et al.Estimation of wheat planting density based on multivariate ridge regression[J].Journal of Agricultural Science and Technology,2020,22(6):72-80.
[22]张沥今,魏夏琰,陆嘉琦,等.Lasso回归:从解释到预测[J].心理科学进展,2020,28(10):1777-1791.
ZHANG Lijin,WEI Xiayan,LU Jiaqi,et al.Lasso regression:from explanation to prediction[J].Advances in Psychological Science,2020,28(10):1777-1791.

相似文献/References:

备注/Memo

备注/Memo:

收稿日期:2021-12-05
基金项目:中国路桥工程有限责任公司科研项目(2020-zlkj-03); 陕西省自然科学基础研究计划项目(2022JM-307)
作者简介:史小丽(1979-),女,工学博士,副教授,硕士生导师,E-mail:glxl@gl.chd.edu.cn。

常用功能

更新日期/Last Update: 2023-07-01