基于案例推理和规则推理融合的跑道侵入事件调查

为提高调查员调查跑道侵入事件的效率和准确度,提出智能辅助决策支持系统的构想。通过辨析基于案例推理和基于规则推理机制的特点发现,两者的缺点可以借助对方的优点予以弥补。在不破坏2种推理技术独立性和完整性的前提下,对其进行融合。在求解各特征属性权重过程中,利用最优传递矩阵的概念改进层次分析法,使之自然满足一致性要求。算例表明,在跑道侵入事件的调查中,无论属性是枚举型特征、数值型特征还是数值区间型特征,相似度计算的方法都是可行可靠的。

Runway Incursions Investigation Based on Integration of Rule-based Reasoning and Case-based Reasoning

In order to improve the efficiency and accuracy of runway incursions investigation made by investigators,conception of intelligent decision support system was put forward.Through the analysis of features between case-based reasoning and rule-based reasoning,it was found that the shortcomings of both can be supplemented by each other's advantage.They can be integrated without any destruction of their respective independence and completeness.In the process of determining the weights of characteristic attributes,AHP was improved by using optimal transfer matrix which can meet the conformance requirements naturally.The example shows that similarity calculation method is feasible and reliable in the investigation of runway incursions without regard to characteristic attributes.