Ideas and possible approaches for ultra-reliability

Ultra-reliability means assured, dependable operation over the required service life with repeatable success. This will be a vital part of any system developed for the NASA space exploration vision. In order to achieve this goal, equipment must be able to survive the stresses of prolonged operation and there has to be a drastic reduction in manufacturing flaws and the human error rate. There will have to be a dedication to the practice of designing in highly robust electrical, mechanical, and electromechanical components into complex systems and special attention to survival of environmental threats such as radiation and micro-meteors. Complex systems present an exceptional challenge because of the large number of potential failure modes as well as the possibilities for system interactions and failure propagation. Failure mode and effects analysis, Event Tree, and fault tree analysis should be applied to the hardware and software as an integrated configuration. In many cases, it will be determined that existing technology has reached its reliability or operating life limit and new technology will be required. Sustained research and development over a period of 20 years or beyond may be needed to develop Ultra-reliable systems on a mass scale. New technology with greater operating life may be able to support the goal of ultra-reliability but that new technology also introduces new reliability problems that will have to be overcome. Resources and established practices for reliability engineering must be strongly maintained as the foundation for achieving ultra-reliability.