Application of a Multiobjective Optimization to Risk-Based Inservice Testing

Abstract

This paper proposes a methodology for applying a multiobjective optimization to risk-based inservice testing with robustness. The multiobjective optimization is applied to solve the tradeoff between maintenance costs and unavailability of a standby system and then assist in determining the robust solution. In order to obtain the most robust solution, a decision-making method for the multiobjective optimization in the viewpoint of robustness is proposed. The risk ranking and revising risk-ranking processes are then used to assist in finding the most optimal surveillance test interval based on risk management. The applicability of the proposed methodology is confirmed by case studies for a standby system of a simplified high-pressure injection system in a nuclear power plant’s pressurized water reactor. The results showed that the proposed methodology provides an effective scheme to achieve the most optimal surveillance test interval based on risk and robustness.