运筹与管理 ›› 2017, Vol. 26 ›› Issue (12): 68-76.DOI: 10.12005/orms.2017.0288

• 理论分析与方法探讨 • 上一篇    下一篇

基于本体关联网络的非常规突发事件案例快速提示方法

李慧嘉1, 贾传亮1, 佘廉2   

  1. 1.中央财经大学 管理科学与工程学院,北京 100080;
    2.国家行政学院 应急管理培训中心,北京 100089
  • 收稿日期:2016-04-07 出版日期:2017-12-25
  • 作者简介:李慧嘉(1985-),男,山东济宁人,副教授,研究方向:应急管理、数据挖掘等;贾传亮(1979-),男,山东莱芜人,副教授,研究方向:运筹学、管理决策分析等;佘廉(1959-),男,湖北人,教授,博导,研究方向包括预警管理、应急决策与指挥等。
  • 基金资助:
    国家自然科学基金项目(71401194,91324203,11131009);中央财经大学“青年英才”培育支持计划资助(QYP1603)

Quick Tips Technology of Unconventional Emergency Based on Ontology Network Analysis

LI Hui-Jia1, JIA Chuan-liang1, SHE Lian2   

  1. 1.School of Management Science and Engineering, Central University of Finance and Economics, Beijing 100080, China;
    2.Chinese Academy of Governance, Beijing 100089, China
  • Received:2016-04-07 Online:2017-12-25

摘要: 由于非常规突发事件具有罕见性、急迫性的特点,要求决策者在具体情境中极速对事件作出决策响应,通过各种非传统和非程式化的方式将事件的危害性降低至最低。为了能够快速地找到最相关的案例并及时地提示指导,本文将非常规突发事件用本体的形式来表示,分析了事件之间的不同类型的逻辑关联及相应程度,并进一步构建了非常规突发事件的本体关联网络。接着,在网络拓扑分析的基础上,我们提出了一种利用交互时间距离CTD(Commute Time Distance)的快速搜索算法,该算法利用谱分析和复杂网络性质分析,能快速找到与新加入节点(新发生突发事件)最相关的案例,从而利用相关措施对紧急危害进行及时地提示和指导。最后,结合青岛“11·22”输油管道爆炸案的实际案例,本文证明了所述方法的有效性并对网络性质进行了分析,结果表明本方法可以高效而方便地应用到各种实际情况中,为非常规突发事件的快速提示提供技术支撑。

关键词: 非常规突发事件, 网络分析, 本体关联, 案例提示, 交互时间距离

Abstract: Due to the rarity and urgency of unconventional emergency, we need to make a decision very quickly by infrequent methods and procedures to minimize the risks of the events. In order to find the most relevant cases and corresponding guidance quickly, in this paper, we first analyze the relationships and the degree of association between events and further develop the ontology-based unconventional emergency network. Then, based on the analysis of network topology, we propose a fast search algorithm using CTD (Commute Time Distance), which employs the spectral analysis and complex network theory. The proposed method can be used to find the most relevant cases quickly with new occurred emergency (the newly added node), and thus performs timely guidance and cautions. Finally, we verify the validity of the proposed method using the Qingdao “11.22” pipeline bombings, and the results demonstrate that the proposed method is very efficient and can be easily applied to a variety of practical situations.

Key words: unconventional emergency, network analysis, ontology correlation, commute time distance, network topology

中图分类号: