Human performance in rail: determining the potential of physiological data from wearable technologies

Fowler, Abigail (2023) Human performance in rail: determining the potential of physiological data from wearable technologies. PhD thesis, University of Nottingham.

PDF (Corrected post viva) (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (7MB) | Preview


This research focuses on how personal data from wearable physiological measures can be used to assess the Mental Workload (MWL) of staff in the rail industry. Automation technologies are being implemented in the rail industry to improve operational performance and capacity. These new technologies are changing the role of staff. This research considers how temporal physiological data present an opportunity to supplement existing workload assessment methods to measure the impact of these technology changes. The research explores how wearable physiological measures could be applied in live operations to collect real-time data with minimal task interference. Whilst the research focuses on railway signallers, the research has implications for other roles in the rail industry and other industries.

The research included three studies and two literature reviews. The initial industry interview study identified the benefit of more continuous data to assess human performance, including successful performance. A detailed review of candidate technologies was then performed solely on physiological measures to extend the knowledge in this area. To assess the potential of physiological measures to provide this continuous data, a simulation study of railway signalling tasks was conducted with an Electrodermal Activity (EDA) wrist strap for alertness and stress and a Heart Rate Variability (HRV) chest strap for uncertainty and increased MWL. The limited application of these measures in rail research provided a suitable research gap for the research to pursue.

The simulation study found physiological data provided visibility of individuals’ personal experience of workload. The interplay of EDA, HRV, task demand and subjective workload over time were visible in the storyboard for each participant. The simulation study provided two key contributions to the thesis. Firstly, EDA identified moments in workload during the task, indicating moments of realisation, and periods of uncertainty, or strain due to time pressure. Such data could be used in staff debriefs to better understand their workload, and tailor training. Secondly, average HRV had a strong negative correlation with average subjective workload. HRV could provide a real time indicator of workload and provide visibility of staff effort to managers.

The final study was an interview and survey study of staff perspectives on the potential use of these measures. This study replaced a live trial which could not proceed during COVID-19 related restrictions. The study found wearable devices suit use in the live operational environment, with the wrist strap rated the most suitable due to low distraction. Trust emerged as a key factor for staff to accept the use of wearables, particularly if named data is shared. Tangible benefits that lead to improvement in operations was identified as one way to build this trust.

An additional contribution of the thesis, drawing on all studies and literature reviews, was to propose a new theoretical perspective on MWL, based on physiological data. A Novelty of Events and Autonomic State (NEAS) model is proposed as a preliminary conceptualisation. It shows how individuals may vary in the impact workload has on their performance and how physiological data may be used to identify this. The concept of Novelty of Events includes aspects of tasks that an individual has not performed before, including those introduced by new technology or procedures. The NEAS model suggests how support in the form of tailored training, or shift breaks, could be used to support improved human performance. Following on from this thesis, a priority for further empirical work would be to trial EDA using a wrist strap that uses a repeated measures approach to determine to what extent individual physiological data changes over time.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Harvey, Catherine
Wilson, Max L.
Sharples, Sarah
Golightly, David
Keywords: Wearable Physiological Measures; Human Performance; Mental Workload; EDA; HRV; Railway Signallers
Subjects: T Technology > TF Railroad engineering and operation
Faculties/Schools: UK Campuses > Faculty of Engineering > Department of Mechanical, Materials and Manufacturing Engineering
Item ID: 73986
Depositing User: Fowler, Abigail
Date Deposited: 05 Sep 2023 12:30
Last Modified: 05 Sep 2023 12:30

Actions (Archive Staff Only)

Edit View Edit View