摘要
为有效预防农村自建房安全事故的发生,对农村自建房安全风险进行识别、评估和管控尤为必要。收集73份事故调查报告,从规划、设计、建造、使用的全过程识别风险因素,构建农村自建房安全风险体系。通过统计每份事故调查报告中风险因素共同出现的次数,建立风险因素共现矩阵,采用Apriori算法挖掘风险因素间的关联规则。为了克服关联规则确定风险因素重要度等级的局限性,综合采用风险因素发生概率及其造成后果严重程度对风险因素重要度等级进行判定,建立风险因素重要度排列矩阵。结果表明:Apriori算法能够用定量的方式表述风险因素之间的关联关系,获得10组风险因素强关联规则,如安全意识淡薄与教育培训不到位同时出现的可能性大,并得到3条具有强关联性的风险因素传播路径;应用改进的风险矩阵得到关键因素10项、次要因素15项、一般因素10项,其中关键风险因素包含设计图纸缺失、违法建设、无施工资质、建筑结构缺陷等。研究结果可为业主和建设主管部门快速进行风险因素重要度评判提供理论支持。
To prevent the occurrence of rural self-built housing accidents, it is particularly necessary to identify, evaluate and control the safety risks of rural self-built housing.Firstly, based on the collected 73 accident investigation reports, the risk factors were indentified from the whole process of survey, design, construction and use of the rural self-built housing.The safety risk system for rural self-built housing was constructed.Then, the co-occurrence matrix of risk factors was established by counting the co-occurrence times of risk factors in each accident investigation report, and the Apriori algorithm was used to examine the association rules between risk factors.Finally, in order to overcome the limitation of association rules to determine the importance level of risk factors, the importance level of risk factors was determined by using the probability of risk factors and the severity of consequences.The matrix of importance ranking of risk factors was established.The result shows that Apriori algorithm can express in a quantitative way to obtain 10 groups of risk factors with strong association rules.For instance, the weak safety awareness and inadequate education and training are likely to occur at the same time.Further analysis was performed to obtain 3 risk factor propagation paths with strong correlations.Moreover, an improved risk matrix method was adopted to obtain 10 critical factors, 15 important factors and 10 general factors.The key risk factors include lack of design drawings, illegal construction, lack of construction qualification, building structure defects, etc.The research results can provide a theoretical support for owners and construction authorities to quickly assess the importance of risk factors.
作者
成连华
曹东强
CHENG Lianhua;CAO Dongqiang(College of Safety Science and Engineering,Xi’an University of Science and Technology,Xi’an 710054,China)
出处
《西安科技大学学报》
CAS
北大核心
2023年第1期73-80,共8页
Journal of Xi’an University of Science and Technology
基金
国家自然科学基金项目(51974238)。
