Meta heuristic intelligent decision support model for uncertainty estimation of life cycle cost in railways rolling stock

Abstract

In Railways rolling stock, Rate of Occurrence of Failures (ROCOF) of repairable components is time varying and has a failure trend, so the combined process of Reliability, Availability, Maintainability and Safety (RAMS) with Life Cycle Cost (LCC) is a stochastic process. Heuristic approaches have been used to design intelligent decision support model to reduce Life Cycle Cost (LCC) and to manage Maintainability to achieve the required Reliability. As heuristic approaches have major drawback of inability to combine both the exploration and exploitation capabilities to reach all regions in the search space, so we propose Meta heuristic approach by hybridization Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). Experimental study proves that our proposed approach significantly avoids this drawback and outperforms the traditional heuristic approaches in reducing LCC while properly maintains the required reliability.