On reliability modeling and analysis of highly-reliable large systems

Modem systems are getting more and more complex, incorporating new technology to meet customer's high expectations. Hardware, software and data communications are integrated to make systems function properly. One example is a critical power system, which provides electrical power to a data center or to semiconductor manufacturing equipment. Reliability techniques of fault-tolerance, true redundancy, multiple grid connections, concurrent maintenance and so forth are applied in design to provide high system reliability. The techniques make the system larger and more complicated in configuration and behavior. Application of modeling tools and analysis methods to such highly reliable, large, complex and repairable systems is discussed in this paper, based on the experience of assessing critical power systems. The use of reliability block diagram plus simulation is recommended as one of the best engineering practices in planning for such large complex repairable systems.