A comprehensive framework for reliability assessment of optical fiber communication systems: From physical degradation mechanisms to network-level evaluation

This paper presents a comprehensive reliability assessment framework for optical fiber communication systems by integrating physical degradation analysis, reliability modeling approaches, and network-level performance evaluation. First, the major failure mechanisms affecting conventional and advanced optical fibers are systematically investigated, highlighting their impact on mechanical integrity and optical transmission quality. Then, various reliability assessment methodologies are reviewed and compared, including statistical models, stochastic degradation models, fracture mechanics-based approaches, environmental reliability models, and advanced monitoring techniques based on optical sensing and artificial intelligence. The proposed framework is applied to the Kongo Central optical fiber network as a practical case study. Conventional reliability models based on exponential and Weibull distributions are evaluated and compared with a Markov-based stochastic approach to analyze degradation evolution, failure probability, and long-term availability. The results demonstrate that the Weibull model provides a more realistic representation of aging-related degradation compared with the exponential model, while the Markov approach enables a detailed description of transitions between healthy, degraded, and failed states by incorporating maintenance effects. The analysis also shows that network reliability is strongly influenced by failure rates, repair duration, and the effectiveness of redundancy strategies. This study provides an integrated methodology for bridging physical failure mechanisms and system-level reliability evaluation, offering valuable insights for predictive maintenance, resilience improvement, and the design of future intelligent optical communication infrastructures.