EEG-based stress evaluation in a ship's bridge simulator based assessment

• Main Author: Gnin, Yeow Dong
• Other Authors: Dimitrios Konovessis
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/68614
• Institution: Nanyang Technological University

Despite advances in technologies that enhance the safety of ships, there is still a relatively high number of casualties within the maritime domain. Research has shown that around 80 percent of incidents happen due to human error. The maritime domain has been relatively slow in recognising the relevance of human factors studies, evident by the limited amount of detailed studies on the subject. Acknowledging this gap in literature, a partnership between Singapore Maritime Academy (SMA), Nanyang Technological University and Fraunhofer IDM aims to introduce and develop a novel method of research using EEG brain state monitoring technology. In this study, experiments are conducted within SMA’s bridge simulator facility to validate the accuracy and viability of EEG technology, and its stress recognition algorithms within the maritime environment. A total of 7 subjects participated in 4 exercises, each designed with an increasingly demanding scenario. EEG data is recorded using off-the-shelf Emotiv Epoch wireless headsets, and then processed using machine learning algorithms to recognise emotion and workload. 2 levels of emotion and 4 levels of workload are recognised, achieving a highest accuracy of 97.14% and 85.71% respectively. Using emotion and workload levels, subject stress levels are estimated. The results undergo statistical analysis and a second-by-second analysis. The study found that the second-by-second analysis of EEG stress data correlates with the simulated situations, and that it was a reliable measure of the difficulty or demand required according to the simulated situation. The result of this study validated the utility of EEG technology in a demanding maritime environment. With EEG technology, researchers and instructors alike benefit from an immensely enhanced picture of a subject’s mental state and an unprecedented ability to analyse the causal effects of certain situations on the cognitive processes of subjects on a second-by-second basis. This opens up many opportunities for researchers to explore and further study human factors, and also for instructors to better assess the abilities and readiness of their students.