Abstract: Traffic blocking is a complicated event that is caused by mutiple factors including meteorological enviromental, road alignment, vehicle status and traffic conditions. As there's a deficiency of research on relevance among these factors, the prevention and control of traffic blocking events have been the difficult issue of road management, especially freeway management. To find out the potential rules and relevance among these factors, this paper proposes a traffic blocking analysis method based on multi-dimensional fuzzy association rules algorithm. In accordance with the need of freeway management,a multi-dimensional attribute model is defined to describe traffic blocking events, based on the relevant national division system and a large number of traffic blocking cases. Then a fuzzy traffic blocking association rules algorithm based on the FCM is set up to mine the potential rules. The practical application of the research proves that the algorithm can effectively support the prevention and management of traffic blocking and thus has bright prospects in traffic blocking analysis.

Key words: traffic blocking, fuzzy association rules, attribute model, FCM

摘要： 交通阻断成因复杂,与气象环境、道路线形、车辆状态以及交通环境等多因素相关。由于缺乏对造成交通阻断相关因素间潜在关联的研究,交通阻断管控一直是公路管理,特别是高速公路管理的难点。本文提出了一种基于多维模糊关联规则的道路交通阻断分析方法,发掘交通阻断的潜在规律和各因素间的关联关系。首先在国家现有相关划分体系和大量交通阻断(事件)案例的基础上,根据道路管理实际需求,建立了交通阻断多维属性模型,然后利用基于FCM的模糊关联规则,挖掘阻断因素的多维属性的依存关系,得到面向道路交通阻断分析的多维模糊关联规则。通过研究成果的实践应用,证明关联规则可以为道路交通阻断预防和管理提供有效支持,在道路交通阻断分析领域有着良好的应用前景。

关键词: 交通阻断, 模糊关联规则, 属性模型, 模糊聚类