地震应急物资需求预测的模糊案例推理技术

地震发生后，及时迅速地应急响应是提升应急救援效率、降低震害损失的重要工作之一，而作为震后灾区应急响应前提和基础的物资需求预测是急需解决的关键问题之一。考虑到震后灾区信息贫乏的特点，引入1种基于模糊案例推理的震后物资需求预测技术。首先，在对已有案例库分析总结的基础上，提取若干影响震后物资需求的地震关键特征属性，通过引入模糊集合的概念，建立地震特征模糊集合；其次，计算新旧案例具体特征属性值对于模糊集的隶属度，为了度量新旧案例之间的相似程度，计算基于新旧案例特征属性权重的修正测度贴近度，贴近度最大者代表新旧案例之间的最佳匹配；最后，使用1个实际案例展示技术的具体应用过程，得到与新案例最近似的已有参考案例，为灾后应急救援提供借鉴。

Study on fuzzy case-based reasoning (FCBR) for demand forecast of emergency material in earthquake

The timely and rapid emergency response after earthquake is one of the important work for promoting emergency rescue efficiency and lowering disaster losses, and the demand forecast of emergency material as the premise and basis of emergency response in post-earthquake disaster area is one of the key problems need to be resolved immediately. Considering the characteristic of poor information in post-earthquake disaster area, a technology for post-earthquake demand forecast of emergency material based on fuzzy case-based reasoning (FCBR) was introduced. Firstly, on the basis of analyzing and summarizing the existing case library, several key attributes affecting post-earthquake material demand were extracted, and the fuzzy sets of the attributes were established respectively through introducing into the concept of fuzzy set. Secondly, the membership degrees of specific attributes' values in new and old cases to the fuzzy sets were calculated. To measure the similarity between new and old cases, the modified measurement similarity degrees were analyzed based on the weights of attributes in new and old cases, and the one with the maximum similarity degree represented the best match between new and old cases. Finally, a real case was applied to present the specific application process of the proposed technology, and the existing reference case most similar to the new case was obtained, which offers a reference for post-earthquake emergency rescue.