Instrumentation & Measurement Magazine 24-4 - 82

A Task Agnostic Mental Fatigue
Assessment Approach Based
on EEG Frequency Bands for
Demanding Maritime Operation
Thiago Gabriel Monteiro, Charlotte Skourup, and Houxiang Zhang
W
hile it is known that one of the major causes of
accidents in the maritime domain is excessive
mental fatigue, objectively assessing mental
fatigue in real-time in maritime operations remains a challenging
and unanswered question. In this work, we aimed to
develop an approach to assess mental fatigue in maritime operators
in real-time and independent of the performed task.
The method involves a simple setup comprised of a wireless
electroencephalogram device, which provides relevant brain
activity data to a mental fatigue assessment algorithm. The
proposed algorithm uses normalized electroencephalogram
energy information to monitor the development of mental fatigue
in maritime operators. We tested our system in a realistic
vessel simulator, and the results showed that it can detect the
increase of mental fatigue levels.
Mental Fatigue Assessment
The poor performance of human operators in human-machine
systems is one of the main contributors to accidents in several
domains, such as driving, airplane piloting and maritime
operations. Factors that can lead to a human operator's poor
performance include improper training, poor communication,
lack of situation awareness, and excessive fatigue. Excessive
fatigue, particularly high mental fatigue (MF), can lead to
performance deterioration due to lack of attention and concentration,
decrease in cognitive capabilities, decrease in reflex
and response time, reduced capacity to evaluate and deal with
risky situations, etc. [1]. Monitoring and assessing MF levels in
human operators can be helpful to reduce accident risks.
In the context of instrumentation and measurement, objective
assessment of MF still an open problem. In order to be
applicable to real-life scenarios, such an assessment system
needs to be nonintrusive from the instrumentation point-ofview,
and reliably assess MF level for cross subject scenarios
from the measurement point-of-view. . These two points are
contrasting, in general, and the trade-off between framework
simplicity and assessment quality needs to be considered [2].
In this light, there is no commercially available system capable
of measuring MF progression reliably.
The most objective and reliable way to monitor MF is
through physiological sensors. In the research literature, there
are reports of systems that make use of multiple physiological
sensors in complex arrangements [3] and in simple and portable
setups [4]. Since low intrusiveness is necessary for an
MF assessment system, the simpler the setup the better. When
focusing on a simple setup that uses only one sensor, electroencephalogram
(EEG) is the most commonly used because it can
detect trends in brain activity that indicate an increase in MF
before any other signal of MF manifests itself [5].
Regarding the reliability of the assessment, the system
should be capable of, distinguishing between several different
MF levels [6] with high precision, so the MF progression can be
properly tracked. This is a complex assessment task in a crosssubject
scenario, as MF manifests differently in different people.
Mental Fatigue and the Maritime
Domain
Most of the reports of MF assessment in the literature are applied
to driving tasks, and although some case studies cover
other fields like aviation [7], very little effort was dedicated
to develop MF assessment systems suitable to the maritime
industry. The maritime industry has extensive regulations
regarding control and management of fatigue [8]. Although
these regulations offer valuable information regarding best
practices to reduce and mitigate the effects of fatigue, they
failed in providing objective ways to measure the development
of MF. An EEG-based MF assessment system could be
used to objectively measure MF progression in maritime operators,
and this information can be used to reduce the risk of
accidents due to human errors.
This research was supported by the project " SFI Offshore Mechatronics " funded by The Research Council of Norway
(project number 237896) and partially supported by the project " SFI Marine Operations " funded by
The Research Council of Norway (project number 237929).
82
IEEE Instrumentation & Measurement Magazine
1094-6969/21/$25.00©2021IEEE
June 2021

Instrumentation & Measurement Magazine 24-4

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 24-4