Integrating human reliability analysis into a comprehensive maintenance optimization strategy

This paper demonstrates the significance of considering human-related characteristics within a system in the overall predictive failure analysis. No maintenance strategy is complete unless it integrates the human operator and the inherent uncertainty into the overall optimization scheme. The inclusion of Human Reliability Analysis (HRA) will expand the traditional focus of condition-based maintenance, thus making it an all-encompassing approach. HRA can capture the uncertainty associated with operator variability for more comprehensive predictive activities within the field of Maintenance Optimization. A well-known model used in failure prediction and reliability analysis is the Proportional Hazards Model (PHM). In many contexts, it makes sense for characteristics related to Human Reliability to be included as new covariates in the PHM. Two systems may start out as identical; but if one utilizes a fully trained crew and the other a completely unskilled crew, the two systems are bound to perform differently with varying reliabilities. The inclusion of Human Reliability in the PHM will make the analysis more complete, resulting in more accurate predictions of failure. This makes a positive financial impact as better maintenance decisions are made. A case study is presented to show the effects of operator skill level on system reliability as a function of time.