A dynamic human reliability model for marine and offshore operations in harsh environments

Human activities, including design, construction, operation, management and maintenance are all a predominant part of the daily tasks in offshore operations. It is not surprising to observe major failures are related to human error, since humans are susceptible to making mistakes. Due to the high level of uncertainty in human activities, predicting all causes of human errors is not an easy process. This may lead to inaccurate results which may affect the overall safety and reliability of marine operations. Evaluation of human endurance during activity on board is a key factor in minimizing risk of human failure. This study aims to examine uncertainties over the time of the marine operation to estimate accurate human reliability assessment. A framework is developed to model the uncertainty of human performance factors by considering a hydrodynamic analysis of the structure along with a subjective analysis of human activities under different weather conditions. Subsequently, a model based on Dynamic Bayesian approach is developed to evaluate the time duration effect on human performance throughout the operation. The developed methodology has been applied to a case study of an offshore vessel storing extracted oil. The framework demonstrates that probability of human failure increases towards the end of its operational days, however the high variation in human reliability is greatly dependent on weather conditions. The present study is able to improve the safety of human life in marine operations by predicting the reliability of performances as a function of time during a specific operation.