Developing a methodology to support the evolution of system of systems using risk analysis

The potential for uncontrolled evolution of member systems is one of the most significant challenges facing SoS (System of Systems) management. Lack of control in evolution makes organisations more inefficient, raises costs, and can affect their ability to adapt quickly to changing circumstance. SoS bring together individual systems to fulfil shared goals, but the systems are likely to have been developed independently, and to different development and operational standards. Member systems retain much of their autonomy, along with their motivations and goals. This increases the risk of decisions relating to evolution being taken unilaterally within member systems, potentially to the detriment of the stability, efficiency and dependability of the SoS. This paper proposes a methodology to support non specialist end users in the identification, organisation and discussion of information required to manage SoS evolution, and uses a modified form of HAZOPS (Hazard and Operability Study) to analyse the associated risks of evolution. The paper evaluates the methodology using a case study based on a recent RAF (Royal Air Force) Nimrod air crash, the official investigation into which showed severe failings in the operation of the Nimrod SoS. The paper concludes that the methodology proposed would provide valuable support to SoS evolution processes, and discusses future directions for the research