Optimization of a bi-objective reliability redundancy allocation problem with heterogeneous components and strategy selection
Ali Najmi, Pardis Pourkarim Guilani
Abstract
As a challenging problem in reliability area, researchers have focused on reliability-redundancy allocation problem (RRAP) to design high reliable systems. Almost all the studies on the RRAP are restricted by two assumptions, using identical components and employing a predetermined redundancy strategy for subsystems. Besides, designing systems with high reliability needs a considerable number of resources, especially financial ones. Consequently, it is crucial to consider the reliability and cost of the system as two conflicting objectives for RRAP. The current study introduces a new mathematical model for a bi-objective RRAP, where the redundancy strategy in each subsystem is considered as a decision variable. Therefore, the mathematical model can select the best strategy for each subsystem from among the three active, standby, or mixed ones. To make the model more realistic, using heterogeneous components in each subsystem is also considered. To have an exact evaluation of the reliability in subsystems with standby and mixed strategies, a Markov-based approach is developed. Finally, an NSGA-II algorithm is utilized to solve the proposed model. The results reveal the superior performance of the proposed model.