Optimization-Based Comparison of Two Reliability Models in Repairable Systems

Abstract

The reliability and performance of repairable systems have long been important concerns in industrial, manufacturing, and engineering environments. Modern production systems are expected to operate continuously with minimal interruptions, making reliability analysis a critical component of system design and maintenance planning. Frequent system failures can lead to production losses, increased maintenance costs, reduced customer satisfaction, and lower overall efficiency. Therefore, understanding the behavior of repairable systems and developing methods to improve their performance remains an essential area of research.Motivated by these challenges, the present study focuses on the analysis of two different repairable system models. The models are evaluated using several widely accepted reliability performance measures, including Mean Time to System Failure (MTSF), Availability, Busy Period, and the Expected Number of Repairman’s Visits. These indices provide valuable information about system behavior by measuring the likelihood of uninterrupted operation, the efficiency of repair activities, and the overall operational effectiveness of the system. Such measures are particularly useful for maintenance managers and reliability engineers when making decisions regarding resource allocation and system improvements.

To perform the analysis, regenerative point techniques are employed, offering a systematic and mathematically sound framework for studying stochastic repairable systems. These techniques enable the derivation of reliability characteristics and performance measures under varying operational conditions. In addition, the study incorporates a metaheuristic optimization approach to determine the optimal values of system parameters, particularly failure and repair rates. Among the available optimization methods, the Moth Flame Optimization (MFO) algorithm is selected due to its effectiveness in solving complex optimization problems and its ability to explore large search spaces efficiently.

The optimized parameter values obtained through the MFO algorithm are used to evaluate and compare the performance of the two repairable system models. Numerical examples and computational illustrations are presented to demonstrate the applicability of the proposed methodology. Comparative analysis of the results highlights the strengths and limitations of each model under different operating conditions. The findings indicate that optimization-based reliability analysis can provide deeper insights into system performance, support informed maintenance planning, and contribute to improved decision-making processes in reliability engineering. Overall, the study emphasizes the importance of combining advanced reliability modeling techniques with modern optimization algorithms to enhance the efficiency, availability, and dependability of repairable systems.