改进的层次分析法及其在自然灾害风险识别中的应用

自然灾害风险识别系统是由众多自然因素和社会经济因素组成的复杂系统 ,其主要问题之一就是如何将有关定性风险因素定量化。为此 ,本文进一步探讨了层次分析法用于自然灾害风险识别系统的具体途径。直接从判断矩阵的定义出发 ,提出了基于加速遗传算法的层次分析法 (AGA AHP)。利用该法 ,可同时确定层次各元素相对重要性的排序权值和进行判断矩阵的一致性检验。理论分析和实例分析的初步结果说明 ,AGA AHP法直观、可行而有效 ,其计算结果精度高于目前常用的基于特征根的层次分析法的相应结果 ,在各种自然灾害复杂系统分析中具有重要的理论意义和广泛的应用价值。

An improved analytic hierarchy process and its application to risk identification of natural disaster

The risk identification system of natural disaster is a complex system, which consists of many natural, economical and social factors. The main problem is how to quantify those qualitative risk factors. For this reason, the application of an analytic hierarchy process to natural disaster identification system is studied in this paper. According to the definition of comparison matrix, an improved analytic hierarchy process based on accelerating genetic algorithm (AGA AHP) is presented. The priority weights of hierarchy elements can be determined and the consistence check of comparison matrix can be done by using AGA AHP. The results from theoretical and example analyses show that AGA AHP is visual, feasible and effective, and the computational precision of AGA AHP is higher than that of the analytic hierarchy process based on exceptional value method. AGA AHP has major theoretic value and wide ranging application in complex system analysis of different natural disasters.