Modeling and Analyzing System Failure Behavior for Reliability Analysis Using Soft Computing-Based Techniques

In recent years, research implications of reliability, availability, and maintainability (RAM) aspects of reliability engineering systems have increased substantially due to rising operating and maintenance costs. For industrial systems, the cost is considered to be the most significant factor and RAM is an increasingly important issue for determining the performance of the system. On the other hand, the information available from the collected databases or records is most of the time imprecise, limited, and uncertain, and the management decisions are based on experience. Thus it is difficult for job analysts to analyze the performance of the system by utilizing these uncertain data. Therefore, the objective of this chapter is to quantify the uncertainties that make the decisions realistic, generic, and extensible for the application domain. For this, an optimization model has been constructed by taking composite measure of RAM parameters called RAM index and system cost as an objective function and solved with evolutionary techniques algorithm. The obtained failure rates and repair times of all constituent components are used for measuring the performance of the system in terms of various reliability parameters using intuitionistic fuzzy set theory and weakest t-norm based arithmetic operations. Performance analysis on system RAM index has also been analyzed to show the effect of taking wrong combinations of their reliability parameters on its performance. The suggested framework has been illustrated with the help of a case.