During a ship life cycle, one of the most critical phases in terms of safety refers to harbor maneuvers, which take place in restricted and congested waters, leading to higher collision and grounding risks in comparison to open sea navigation. In this scenario, a single accident may stop the harbor's traffic as well as incur in patrimonial damage, environmental pollution, human casualties, and reputation losses. In order to support the vessel's captain during the maneuver, local experienced maritime pilots stay on board coordinating the ship navigation while in restricted waters. Aiming to assess the main factors contributing to human errors in pilot-assisted harbor ship maneuvers, this work proposes a Bayesian network model for human reliability analysis (HRA), supported by a prospective human performance model for quantification. The novelty of this work resides into two aspects: (a) incorporation of harbor specific conditions for maritime navigation HRA, including the performance of ship's crew and maritime pilots and (b) the use of a prospective human performance model as an alternative to expert's opinion for quantification purposes. To illustrate the usage of the proposed methodology, this paper presents an analysis of the route keeping task along waterways, starting from the quantification of human error probabilities (HEP) and including the ranking of the main performance shaping factors that contribute to the HEP.

In the oil sector, analysis, evaluation, and management of risk are vital, considering the accidents potential severity with respect to human life, environment and property. Since most accidents in this sector include human factors, and given the lack of suitable tools for its consideration along the systems life cycle, especially during design phase, the development of models dedicated to human factors in risk analysis is essential. In this context, a technique for early consideration of human reliability (TECHR) was designed for developing a prospective human performance model, which can be exploited in the system design phase and which can be updated along the system life cycle. TECHR is based on the use of expert opinion in relation to systems that operate or have operated in recent years to obtain estimates of the probabilities of the various types of human error which may occur during the performance of a specific action. This paper presents the application of a prospective human performance model—obtained by TECHR—in the study of an oil tanker operation, focusing on human factor quantification in scenarios of collision. In this work, the actions presented in a previous fault tree—for vessel operation—are quantified considering the mentioned model, and the results are discussed in view of the previous results of this fault tree that used the human error probabilities (HEPs) presented in the technique for human error rate prediction (THERP), allowing the comparison of the results obtained by the THERP with the results obtained by the TECHR.

Data scarcity has always been a significant challenge in the domain of human reliability analysis (HRA). The advancement of simulation technologies provides opportunities to collect human performance data that can facilitate both the development and validation paradigms of HRA. The potential of simulator data to improve HRA can be tapped through the use of advanced machine learning tools like Bayesian methods. Except for Bayesian networks, Bayesian methods have not been widely used in the HRA community. This paper uses a Bayesian method to enhance human error probability (HEP) assessment in offshore emergency situations using data generated in a simulator. Assessment begins by using constrained noninformative priors to define the HEPs in emergency situations. An experiment is then conducted in a simulator to collect human performance data in a set of emergency scenarios. Data collected during the experiment are used to update the priors and obtain informed posteriors. Use of the informed posteriors enables better understanding of the performance, and a more reliable and objective assessment of human reliability, compared to traditional assessment using expert judgment.