Knowledge Management for Fault Tree Analysis Based on Quantity Dimension Indexing

In this paper, the authors propose computerized support for fault tree analysis (FTA) based on a new design knowledge management approach called quantity dimension indexing. FTA is a method of analyzing and visualizing the causes of fault events by expanding a fault event hierarchically to its possible cause events and constructing a tree diagram representing the entire structure of the problem. When a designer finds or encounters a problem during a product design and development process, an effective way of ensuring the security and safety of the product is to identify all the possible causes of the problem by FTA and fix them. Although FTA is an effective method, it is not easy for a designer to construct a complete fault tree without any misunderstanding or oversight. A promising approach for supporting FTA is to utilize a computerized knowledge management method. Although many knowledge management techniques for literal expression have been developed, they are not necessarily suitable for managing the engineering design knowledge of physical phenomena. To solve this problem, the authors propose a new design knowledge management approach called quantity dimension indexing and computerized support for FTA such as the verification of consistency of a fault tree and fault tree construction advice. By analyzing fault tree examples based on actual design activities in a company, the possible feasibility and future promise of the proposed approach are indicated.