BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260523T082641EDT-4328Z5URVe@132.216.98.100
DTSTAMP:20260523T122641Z
DESCRIPTION:Abstract\n\nHuman automation teams are increasingly prevalent i
 n domains such as aviation\, autonomous vehicles\, healthcare\, and manufa
 cturing. In such systems\, the cognitive states of humans\, including ment
 al workload and fatigue\, play a pivotal role in determining the joint per
 formance of the team.\n\n \n\nIn this thesis we focus on the dynamic natur
 e of the human cognitive state\, which exhibits temporal correlation. Subj
 ective measures of mental workload can be obtained using standard question
 naires like the NASA-TLX\, however\, their administration is often impract
 ical as it interferes with the primary tasks of the human operator. Theref
 ore\, it is of interest to estimate these subjective measures from less in
 trusive observations. Evidence suggests that mental workload is a dynamic 
 process\, so incorporating historical measurements could reduce its estima
 tion error. Additionally\, the estimation of operator performance in human
  automation teams is essential in optimizing task effectiveness and facili
 tating efficient resource allocation. We present and compare different dyn
 amic schemes to estimate an operator’s performance on classification tasks
 \, i.e.\, classification accuracy\, and her subjective ratings on subscale
 s of the NASA-TLX questionnaire\, which measure mental workload across mul
 tiple dimensions. These schemes differ in the information available for es
 timation. We test these schemes on data collected from a scenario where a 
 human and an automation perform a series of classification tasks for simul
 ated mobile objects. Our analysis of the interaction data and the estimati
 on schemes indicates that employing dynamic estimation leads to decreased 
 estimation errors for certain NASA-TLX subscale ratings.\n\n \n\nWe next a
 ddress the problem of allocating decision tasks\, via decision referrals f
 rom the automation to the human\, in a context where human performance dep
 ends on workload and a dynamically evolving level of fatigue. To model thi
 s\, we formulate the decision referral problem as a MDP\, allowing us to e
 xplicitly account for the changing state of human fatigue and its impact o
 n decision-making performance. To solve the MDP\, we propose an approximat
 e dynamic programming approach\, which yields a fatigue-aware decision ref
 erral policy. We evaluate the performance of this policy\, which considers
  fatigue dynamics\, against a task deferral policy that bases its decision
 s solely on instantaneous fatigue levels. We then analyze the robustness o
 f the fatigue-aware policy.\n
DTSTART:20250321T180000Z
DTEND:20250321T200000Z
LOCATION:Room 603\, McConnell Engineering Building\, CA\, QC\, Montreal\, H
 3A 0E9\, 3480 rue University
SUMMARY:PhD defence of Raihan Seraj – Dynamic modeling of mental workload a
 nd fatigue-aware decision referrals in human-automation teams
URL:/ece/channels/event/phd-defence-raihan-seraj-dynam
 ic-modeling-mental-workload-and-fatigue-aware-decision-referrals-human-364
 234
END:VEVENT
END:VCALENDAR