International Journal of Reliability, Quality and Safety Engineering

Multistate start-up demonstration tests are proposed and analyzed using a Markov Chain approach. The new demonstration tests are generalizations of the CSTF, TSTF, and CSCF start-up demonstration tests discussed in the literature that have the important practical advantage of allowing for early termination when the quality of successful start-ups is exceptional. When the termination results in acceptance of the equipment, the use of a multi-state startup demonstration test results in an increase in the probability of an early termination of as much as. 35 over the optimal binary start-up test in cases studied. The Markov Chains methodology is used in the probabilistic analysis. Practical guidance on choosing tests, estimation, and comparisons of the various criteria are studied. The Markov Chain approach is then extended to study the non-i.i.d. case.

The reliability importance of one or more components when another component is assumed to be working/non-working is measured by Conditional Marginal Reliability Importance (CMRI) and Conditional Joint Reliability Importance (CJRI) respectively. We consider two systems viz the series-in-parallel and series-parallel. The expressions for CMRI and CJRI are derived for both the systems when the components are independent but not identically distributed. It is shown that the sign of the joint importance of three components and Conditional Joint Importance (CJI) can be determined using Schur-convexity (concavity) of the reliability function. The difference in the reliability functions of two coherent systems with n ≥ 3 statistically independent and with dependent components is derived. It is shown to be measured by their covariance, the JRI and the CJRIs. CMRIs and CJRIs of a phased type electronic system and a bridge structure are worked out.

The Markov Chain Modular (MCM) approach is proposed in this paper in order to solve part of the failure-state dependency problem. The MCM approach completely avoids the failure-state dependency problem by avoiding the combinatorial modeling. To quantitatively assess safety, a new Markov chain modeling technique is developed to represent an m + 2 state homogenous Markov chain model using a three-state Markov model. The transition rate functions of the three-state Markov model can be determined by the transition rates of the m + 2 state Markov chain model. Given a series system has N modules and each module has O(m) operational states, the MCM approach reduces the operational states to O(N × m²) as opposed to O(m^2N) by using the traditional Markov chain model.

Surveillance Test Interval is an important parameter for the standby systems of a plant with respect to its availability, cost and other issues. Stand-by systems are not required during normal operations but are essentially required when demanded by the plant operations. Stand-by systems are tested and maintained periodically in order to ensure their serviceability.

In this paper an attempt has been done to optimize the unavailability, cost and manrem consumption with respect to Surveillance Test Interval for the standby systems of nuclear plants. In this work a novel approach has been introduced where Real Parameter Genetic Algorithm (GA) has been used for the multi-objective optimization problem in hand. Application of Genetic Algorithms in similar problems has been reported in literature. But the approach proposed in this paper differs from the exixting methods significantly. In this work real-parameter GA has been used which makes the algorithm simple by not having the overhead of encoding and decoding of solutions. More-over a multi-objective optimization method has been proposed that not only takes care of optimizing the unavailability but also cost and manrem consumption. Here we have mainly concentrated on the Emergency Core Cooling System of a Research Reactor, but the same idea can easily be extended for the other standby systems.

We develop a computer design process that uses an architectural approach, involving the definition and analysis of sequences of functions. This approach allowed us to configure a system at a high level, and provides a mechanism for predicting performance, reliability, availability, and security. A key feature of the design approach is sequence and system complexity that we found to be a good predictor of system properties, such as reliability. Both non-concurrent and concurrent processing, using multiple parallel processors, methods are evaluated with respect to performance, reliability, availability, security, and cost. We evaluated the tradeoffs between performance and cost, and found that a two-processor system provides a good balance between performance and cost. Our major contribution to the field of computer design is our innovative approach of using sequences as the basis of design and the application of complexity metrics as a predictor of system attributes, an approach we did not find in an extensive review of the literature.