Risk assessment of real time digital control systems

This paper describes stochastic methods for assessing risk in integrated hardware and software systems. The methods assess evaluate availability, outage probabilities, and effectiveness-weighted degraded states based on data from measurements with a specified confidence level. System-level reliability/availability models can also identify the elements where failure rate, recovery probability, or recovery time improvement will provide the greatest benefit. The validity of this approach is determined by the extent to which the system failure behavior conforms to a stochastic process (i.e., random, non-deterministic failures). Evidence from large studies of other high availability computer systems provides substantial evidence of such behavior in mature systems. The approach is limited to the systems with failure rates higher than 10^-6per hour and the availability below 0.999999, i.e., below safety grade. To assess safety critical systems, the risk assessment method described here can be used as an adjunct for other approaches described in various industry standards that intended to minimize the likelihood that deterministic defects are introduced into the system design