节点文献

基于贝叶斯网络的建筑施工安全事故致因研究

Research on Causes of Construction Safety Accidents Using Bayesian Network

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 薛楠楠张建荣张伟赵挺生

【Author】 XUE Nannan;ZHANG Jianrong;ZHANG Wei;ZHAO Tingsheng;School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology;

【通讯作者】 张伟;

【机构】 华中科技大学土木与水利工程学院

【摘要】 施工环境复杂多变且影响安全管理的因素众多,因此有必要对导致施工安全事故的因素进行分级管控,从而提高安全管理效率、降低事故发生概率。基于系统分析和事故致因理论,将建筑施工安全事故致因作为一个系统,划分为6个子系统和33个因素。以1484份事故报告为样本,结合关联规则和专家知识确定因素之间关联,建立施工安全事故致因贝叶斯网络模型。通过诊断推理和敏感性分析,得到施工安全事故发生的最大可能性路径,确定了引发施工安全事故的9项关键因素和4项敏感因素。并以一起模板支架坍塌事故为例,对比事故致因的实际发生情况与贝叶斯分析结果,验证了施工安全事故致因贝叶斯网络模型的实用性和分析结果的合理性,为安全管控和事故预防提供依据。

【Abstract】 Many influencing factors were involved in construction projects with a complex and changeable construction environment. Therefore, it is necessary to identify and control these factors aiming to avoid accident occurrence. Based on the system thinking and accident causation theory, construction accident cause was considered as a system and decomposed into 6 subsystems and 33 factors. A sample of 1484 accident reports was collected and association rules of accident causes were refined, after that the topological structure and Bayesian network parameters were determined. The diagnostic reasoning and sensitivity analysis were conducted, from which the most possibility path of construction safety accidents was found, and the 9 critical factors and 4 sensitive factors were identified. Finally, a real formwork collapse accident was adopted as a case study, which proved the practicality of the established Bayesian network model and the reasonability of analysis results, and provided a basis for safety control and accident prevention.

【基金】 国家重点研发计划(2017YFC0805500);国家自然科学基金(51308240)
  • 【文献出处】 土木工程与管理学报 ,Journal of Civil Engineering and Management , 编辑部邮箱 ,2021年04期
  • 【分类号】TU712.4
  • 【被引频次】6
  • 【下载频次】926
节点文献中: