The reliability and robustness are critical aspects for engineering systems. In a complex system, usually multiple functions need to be achieved. In such cases, there exist optimal mappings between functions and physical components. In this paper, a new approach, Axiomatic Fault Tree Analysis (AFTA), is proposed to improve the system reliability and robustness simultaneously. In AFTA, system robustness is improved by using Axiomatic Design (AD) to decouple the system to an extent where a group of functions have a corresponding relationship with only one physical component. Moreover, Fault Tree Analysis and Axiomatic Design have been integrated in AFTA by using two new indices indicating the importance degrees of functions and physical components in a reliability sense, which provides a guideline for reliability improvement. A case study of a complex system is used to illustrate the procedure and applicability of AFTA. Theoretical calculations are used to verify AFTA for non-repairable systems, and multi-agent based simulations are for repairable systems. Both the theoretical and simulated results show that the system reliability has been improved significantly without large additional cost.

This content is only available via PDF.
You do not currently have access to this content.