面向应急决策的极端洪水关键情景推理研究  

Study of Extreme Flood Key Scenario Reasoning for Emergency Decision

作　　者：尹洁[1,2] 施琴芬[3] 李锋[2] Yin Jie;Shi Qinfen;Li Feng(Business School,Hohai University,Nanjing 210098;School of Economics and Management,Jiangsu University of Science and Technology,Zhenjiang 212003;Suzhou University of Science and Technology,Suzhou 215009)

机构地区：[1]河海大学商学院,南京210098 [2]江苏科技大学经济管理学院,镇江212003 [3]苏州科技大学,苏州215009

出　　处：《管理评论》2019年第10期255-262,共8页Management Review

基　　金：国家自然科学基金青年项目(71303074)。

摘　　要：有效的情景推理能够帮助应急决策者在极端洪水关键情景的高压力下制定合理的对策,而现有案例推理方法无法解决数据类型多样、历史样本量小等难题。本文基于知识元理论划分并界定极端洪水关键情景,结合知识元间关系、专家知识,构建情景检索信度规则库体系及情景相似度评估的信度规则库,通过证据推理方法融合信度规则,完成极端洪水关键情景检索及推理,为极端洪水关键情景的应对提供最合理的决策支持。以淮河流域蒙洼蓄洪区启用情景为例,验证了极端洪水关键情景推理方法的适用性与可行性。Effective scenario reasoning can help emergency decision makers work out reasonable solution under high pressure of the extreme flood key scenario,but the existing method of CBR could not address the difficulty caused by multiple data types and insufficient historic samples. Extreme flood scenario is defined and represented generally based on knowledge element by scenario division,and belief rule-base for scenario reasoning is constructed based on relation of knowledge elements and expert knowledge. Based on belief rule-based inference methodology by evidential reasoning approach,extreme flood key scenario retrieval and reasoning is completed,which could improve the effectiveness of decision-making by historical emergency solution for extreme flood key scenario emergency management. Taking the scenario of Mengwa flood storage areas starting in Huaihe Valley as an example,the applicability and feasibility of this method is proved.

关 键 词：极端洪水 关键情景 知识元 证据推理 信度规则库 

分 类 号：G63[文化科学—教育学]