【作者】 闻千；

【导师】 李宗平；

【作者基本信息】 西南交通大学 ， 交通运输规划与管理， 2013， 硕士

【摘要】 目前我国城市轨道交通进入了一个高速发展期,多数城市的轨道交通都由以前的单线运行转换成了多线、多制式运行的网络化运营局面。网络化运营条件下,城市轨道交通网络结构复杂,运营制式多元,发生突发事故后,事故影响面大,后果比较严重,这给应急指挥管理带来很大难度。因此科学合理的应急指挥管理方法,对城市轨道交通预防、减少突发事故,降低突发事故伤害具有重要意义。进行城市轨道交通应急指挥管理评价是完善应急指挥管理方法、提高应急指挥管理能力的基础。通过建立精准的评价模型,评估城市轨道交通应急指挥管理能力,发现城市轨道交通应急指挥管理过程中存在的问题,为形成高效的应急指挥管理机制提供理论依据。本文对国内外应急能力评价的现状进行了研究,发现目前国内外的大部分学者对于应急能力评价指标都是通过总结借鉴前人的研究成果或者根据自己的主观想法进行构建的,这样得到的指标体系缺乏说服力,其合理性也有待考证。鉴于此,本文基于事故分析构建城市轨道交通应急指挥管理评价指标体系,介绍了城市轨道交通诞生以来的150年间世界各国城市轨道交通系统发生的各种典型突发事故的基本情况,并对事故发生的原因、处置过程进行了深入分析,得到了城市轨道交通应急指挥管理的影响因素,以影响因素为基础初步建立了评价指标体系,得到4个一级指标(分别为人员素质、设施设备、管理机制和社会控制)和30个二级指标。基于事故分析得到的评价指标是全面但粗糙的,为了体现模型的精准性,需要对初步得到的评价指标体系进行约简。粗糙集理论具有这种去除冗余指标,提取核心指标的功能。基于粗糙集的属性约简性,本文建立了城市轨道交通应急指挥管理评价指标约简模型,基于粗糙集的信息量和属性重要性原理,建立了指标权重确定模型。选取我国城市轨道交通运行年限相对较长的7座城市的城市轨道交通(北京地铁、上海地铁、广州地铁、深圳地铁、香港地铁、重庆轻轨、成都地铁)作为研究对象,通过实地调研及专家咨询等方式给初步建立的评价指标体系进行赋值,建立了城市轨道交通应急指挥管理评价信息系统,结合指标约简模型和权重确定模型,得到了21个最佳约简指标及其综合权重。最后采用线性加权法对各个指标进行加权计算,从而得到各评价对象的综合评价结果。更多还原

【Abstract】 At present, urban rail transit in our country has entered a high-speed development period. The majority of city’s urban rail transits turn to a new network operating situation under the old single-line operation transformed into a multi-line multi-format one. Under this condition, the complexity of network structure,the multiplicity of operating standard besides the big area of infection and the serious consequence after a sudden accident occurred. These bring many difficulties to the emergency command management. As a result, the scientific and reasonable method of emergency command management for prevention of urban rail transit reducing sudden accidents and decreasing accident damage is of great significance.An emergency command management evaluation of urban rail transit is a nice precondition for establishing, perfecting and improving the ability of it. Through the establishment of accurate model and evaluation of urban rail transit emergency command management ability, finding the problems existing in the emergency command management process, repeatedly to improve and perfect, then getting a better emergency command management ability in order to reducing sudden accidents and decreasing accident damage.Regarding to the domestic and foreign emergency capability evaluation, the author has researched in depth. However, the research result showed that nowadays most of domestic and foreign index mostly basing on the research results of predecessors or subjective thoughts. By this way, the index system certainly is lacking in convincing, its rationality is yet to be proven. In view of this, it was introduced that the basic situation of a variety of typical accidents happened in all the countries around the world since the birth of urban rail transit for150years. Based on the influencing factors which got from analyzing the cause of accident and the disposal process deeply, the preliminary evaluation index system was established including4first-level indexes (personal quality, facilities, management mechanism and social control) and30second-level indexes.As the evaluation index based on accident analysis is comprehensive but rough, it’s necessary to simplify preliminary evaluation index system for reflecting the accuracy of model. Rough set theory owns the functions that removing redundant indexes and extract the key one. Based on the simplicity attribute of rough set, the selection model of urban rail transit emergency command management evaluation index was established and based on information and impotence of attribute principle of rough set, the model which determined by index was also established. In this article7cities whose operating time of urban rail transit are longer were selected as research objects (Beijing subway, Shanghai subway,Guangzhou subway, Shenzhen subway,Hong Kong subway,Chongqing light rail,Chengdu subway). By means of assigning the preliminary established evaluation index system which got form field survey and expert consulting, the information system of urban rail transit emergency command management was established. Combining index selection model and determining weight model, we got21the most simple indexes and the comprehensive weight. So each index was calculated by using the linear weighting method, and obtained the comprehensive evaluation results finally.更多还原