Over the last ten years economic pressure have prompted the maritime industry to reduce crew size on vessels. Despite a parallel increase in automation, this development has had a significant impact on shipboard operations, where crewmembers today are generally required to cope by taking on more tasks and working longer hours. Moreover, the International Maritime Organization reports that more than 75 percent of ship accidents worldwide are due to human error. Within this context, the industry has reached a critical point where operational safety can only be achieved on the basis of a better understanding of human factors.
This thesis focuses on Royal Australian Navy patrol boat operations and investigates human error related problems, specifically factors such as situation awareness, fatigue, workload and stress. These factors were examined over four studies. The first three studies aimed to be exploratory in nature. In the first study human error and recovery data were drawn from incident reports using such data gathering tools as the Critical Incident Technique and the Critical Decision Method. The findings in this study showed that a number of problems identified on board are related to cognitive errors. These can stem from performance influencing factors such as situation awareness, fatigue, workload and /or stress. The second study focused on one aspect of human error and addressed the issue of lack of situation awareness in the maritime domain. The findings of this study demonstrate that factors such as workload and fatigue have important implications on situation awareness and safety. To probe deeper into this, a questionnaire survey was utilised for the third study. This focused on fatigue related issues. Findings from the third study led to concerns over the negative effects of time, workload, and stress levels on crewmembers fatigue during onboard operations. The literature review for this thesis revealed very limited research examining the effects of potential risk factors on fatigue. Hence, the fourth study attempts to address this research gap by addressing a number of hypotheses. Hierarchical Linear Modelling (HLM) was used to evaluate the antecedents of fatigue from multiple levels of analyses. The results indicate that fatigue changed throughout the day and across days and the effects of workload and tense arousal on fatigue changed across days. These results contribute to our theoretical understanding of dynamic relationships among fatigue and associated risk factors. The implications of this work are discussed in the final chapter of the thesis.