Inspection and maintenance optimization for heterogeneity units in redundant structure with Non-dominated Sorting...

Area of Science:

• Reliability Engineering
• System Safety Analysis
• Stochastic Processes

Background:

• Redundant systems enhance reliability by maintaining function after unit failure.
• Traditional models assume uniform unit aging and shock resistance, which is often unrealistic.
• Individual unit shocks necessitate tailored inspection strategies for safety.

Purpose of the Study:

• To address stochastic dependencies in redundant systems using a novel activation sequence.
• To propose an adaptive system-level inspection policy prioritizing units with worse states.
• To optimize system unavailability and maintenance costs through advanced algorithms.

Main Methods:

• Introduction of a novel activation sequence to model stochastic dependencies.
• Development of an adaptive inspection policy that prioritizes degrading units.
• Utilization of Monte Carlo simulations to estimate system unavailability and maintenance costs.
• Application of the Non-dominated Sorting Genetic Algorithm III (NSGA-III) for multi-objective optimization.

Main Results:

• Demonstrated contradiction between minimizing system unavailability and maintenance costs.
• Identified optimal trade-off solutions between system unavailability and maintenance costs.
• Validated the effectiveness of the proposed adaptive inspection policy.

Conclusions:

• The novel activation sequence and adaptive inspection policy effectively manage redundant system reliability.
• The NSGA-III algorithm provides valuable insights for practitioners in balancing system performance and cost.
• This research offers a more realistic approach to inspecting and maintaining complex redundant systems.