Fault Tree Modeling for Redundant Multi-Functional Digital Systems

Abstract

A digital system is multi-functional and it may cause a risk concentration if it is installed without any redundancy. In order to reduce this risk of a concentration, multiple redundancies are applied for the safety-critical systems. Simplified common cause failure (CCF) modeling technique is commonly used for a practical reason when the target system consists of many redundant components. This technique enables us to avoid an impractically large number of CCF events for a system fault tree. One of its demerits is that it may hide important information regarding a systems status. Condition-based human reliability assessment (CBHRA) method which is a practical method to develop a more realistic fault-tree model with a consideration of the various conditions endured by a human operator uses the system status information for post-processing of minimum cut sets. Since the CBHRA post-processing should be performed based on the status information of the sensors and the signal-processing equipment and noting that the simplified CCF modeling technique may cause the loss of some information, in order to perform a more realistic analysis, we have to overcome this possible information loss. In this study, we also present a case study of a fault tree modeling which adopted both the CBHRA method and the simplified alpha factor method.