基于偏好信息的铁路应急预案多阶段群评估研究

Multi-stage Group Evaluation of Railway Emergency Plans Based on Preference Information

铁路应急预案评估是铁路应急管理体系的重要组成部分,针对现有研究存在的不足,从组织指挥机构及职责、应急保障措施、宣传培训与演练、预防预警、应急响应、后期处置等6个方面构建针对性较强的评估指标体系,提出基于偏好信息与指标关联的多阶段群评估方法以确定最佳预案。首先,针对各评估阶段,组织评估专家以互补判断矩阵的形式给出各指标预案的偏好信息;其次,检验并改进互补判断矩阵的一致性水平以及群体一致性水平,基于组合赋权法确定专家权重,以集结专家的偏好信息;然后,利用ANP法获得指标关联权重,集结所有指标的群体偏好信息;最后,构建最小偏差模型求解阶段权重,集结各阶段的群体偏好信息,以此确定候选预案集的优劣排序。案例表明,本文提出方法具有可行性与有效性。

The evaluation of railway emergency plans is a key part of the railway emergency management system.With respect to the inefficiency of the existing study of the evaluation of railway emergency plans,a novel evaluation index system of the railway emergency plans was proposed by including commanding organization and its responsibilities,emergency safeguard measures,publicity,training and drilling,prevention and early warning,emergency response,handling in the aftermath of an accident.In order to select the best railway emergency plan from the candidate plans,an index-associated approach for multi-stage group evaluation of railway emergency plans based on preference information was put forward.Firstly,for each index of each evaluation stage,preference information which reflects pros and cons between the plans was given by experts in the form of complementary judgment matrix.Secondly,the complementary judgment matrix was tested and modified until all preference relations reached an acceptable predefined individual consistency and group consensus level.Based on combination assigning method,the weight of experts was determined to gather the preference information of all experts.And then the ANP method was used to obtain index associated weight,to gather group preference information of each index.Finally,the minimum deviation model was developed to solve the weight of each stage to gather group preference information of each stage,to determine the order of priority of candidate emergency plans.The example showed that the method is feasible and effective.