Computational intelligence in reliability and maintainability engineering

In this paper the basics of reliability and maintainability modeling, prediction and optimization problems using stochastic models are briefly reviewed (for non-repairable and repairable systems). As an alternative to classical methods based on stochastic models, computational intelligence techniques such as neural networks and fuzzy systems as well as evolutionary computing, artificial immune systems and swarm intelligence are introduced. Classical methods, neural networks, evolutionary computing and immune algorithm are followed by examples demonstrating their applicability to reliability modeling, analysis and optimization. This is a fairly new research area and it has a great potential to support engineers on solving problems such as modeling, analysis and optimization of real-world industrial systems.