基于知识元的应急决策知识表示与生成方法

Method on emergency decision knowledge representation and generation based on knowledge-unit

非常规突发事件应急决策是一类应急状态下的复杂系统决策问题,快速生成应急决策知识是解决这类问题关键.本文从不同科学域的共性整体特征出发,构建应急决策知识元模型,为多领域、多学科显性知识和隐性知识的表示与融合提供支持.在此基础上,本文首先基于知识元模型生成个体知识,然后通过GRA-T(grey relation analysis and TOPSIS)方法确定决策主体考虑问题的权重,再使用WGRG(weight grey relation grade)融合算子将专家知识整合成系统性的应急决策知识,为保证应急决策的及时性、科学性和有效性提供知识支持.最后,通过数值算例和地铁火灾突发事件案例分别验证GRA-T和WGRG方法的优越性以及应急决策知识表示和生成方法的实用性和科学性.

Unconventional emergency decision-making is a complex system, the key to resolve this problem is rapidly generating emergency decision knowledge. The paper built emergency decision knowledge-unit model on the strength of the common characteristics of different scientific domains, which gives support to integrate with multi-disciplinary tacit and explicit knowledge. On this basis, the paper generated experts knowledge by integrating the experts＇ experience with emergency knowledge, then built the grey relation analysis and TOPSIS （GRA-T） method to determine the weight of decision goals and used weight grey relation grade （WCRG） operator to generate emergency decision knowledge, which help decision-makers make decision with timeliness, scientificity and effectiveness. Finally, example of a numerical computation was presented to show the advantages of GRA-T and WGRG, and example of the subway fire emergency was presented to prove the emergency decision knowledge generation is applied and scientific.