应急决策是基于灾情信息的,而灾情信息往往是多源、不确定性的。信息高度缺失和模糊是巨灾下应急决策的主要特点之一。如何在多源不确定情况下对现有信息进行处理,满足应急决策支持的特定需求是文章的主要研究问题。认为融合同一情境的多个数据集信息是改善应急决策目标识别的一种较为有效的方式,即将描述同一灾情的不同方面信息融合处理以实现灾情态势评估,为应急决策提供基础。基于这一思想,构建了应急信息的融合体系框架,给出了主要的处理流程,并针对流程中决策级融合处理过程中的信息更新和多辨识框架融合问题给出了主要的算法。该方法能够实现灾情态势快速识别、评估的目标。
Emergency decision-making is based on disaster information, which is always multi-source and uncertainty. The ambiguity and lack of information is one of the main characteristics of emergency decision-making under the catastrophe. How to process the existing information and satisfy the specific needs of emergency decision-making in the situation of multi-source and uncertainty is the main research problem of this paper. The paper believes that the fusion of multiple data sets in the same context is an effective way to improve the target recognition of emergency decision-making, namely, to describe the integration of different information for one disaster to realize the disaster situation assessment, which provides the basis for the emergency decision-making. Based on the ideas above, the paper builds the fusion system framework of emergency information, puts forward the main processing procedures, and proposes the main algorithm for information updating and multi-recognition framework fusion in the process of decision-level fusion. This method can achieve the goal of recognizing and evaluating the disaster situation.