An FMEA-based methodology for the development of control software reliable to hardware failures

In automation systems, a high number of faults is induced by hardware failures. Their control software can be utilized to mitigate this problem by making it detect and manage the different failure events that may occur in the system. However, control software design methodologies have mainly focused on the system nominal behavior, marginally consider the generation of software reliable to hardware failures. In response to this challenge, this paper presents a methodology for the development of reliable automation systems which integrates the following tools: (i) Failure Mode and Effect Analysis (FMEA): to identify the different failure modes, and the strategies for their detection and management; (ii) AutomationML: to model the hierarchy and interfaces of automation system's components; (iii) Virtual Commissioning and Fault Injection: to assess - before system deployment - the reliability of the control software in the presence of hardware failures. Through its application to a case study, it is demonstrated that the methodology enables the identification of failure modes, the elicitation of requirements for their detection and management, and the generation of control software reliable to the identified failure modes.