Am Montag, den 08. April 2024, um 16:15 Uhr hält

Jannike Illing
Universität Oldenburg

im Rahmen ihrer beabsichtigten Dissertation einen Vortrag mit dem Titel

Supporting Time-Critical, Parallel and Spatially Distributed Tasks Using
Augmented Reality

Der Vortrag findet hybrid statt:
OFFIS, Escherweg 2, Raum F 02 und https://bbb.offis.de/rooms/jeo-3kf-jtu-476/join

Abstract:
While many tasks are becoming increasingly automated, there are still a significant number of tasks that require
human intervention - especially when their complexity requires adaptability beyond the current capabilities of
automated systems. An example of this is assembly, where multiple complex tasks are distributed across multiple
locations and often still rely heavily on manual labor, even as automation progresses. Tasks with systematic processes
often have manageable cognitive demands when performed individually and without time pressure. In many
real-world work scenarios, however, such tasks are performed under time pressure. In addition, multiple tasks
must be performed simultaneously, which may be distributed across multiple workstations. The complexity of
such tasks is therefore significantly increased by the inherent time criticality, simultaneous execution, and spa-tial
distribution.
This dissertation investigates how Augmented Reality (AR) assistance should be designed to reduce the complexity
of time-critical tasks. We follow the Human-Centered Design approach, where the potential users are the starting
point and benchmark for the design during development and evaluation. To this end, we first analyzed the
context of use through observation and expert interviews in order to derive requirements for the support options.
Three research questions were derived from the analysis, which will be investigated in the dissertation: (RQ1)
How can time-critical and parallel tasks be visualized?, (RQ2) How can time-critical, parallel and spatially distributed
tasks be guided?, and (RQ3) How can time-critical, parallel and spatially distributed tasks be monitored?
Our results show that AR reduces the complexity of time-critical, simultaneous, and spatially distributed tasks and
supports users in their execution. We show that AR instructions are preferred over traditional (paper) instructions
for time-critical tasks and lead to shorter execution times. We also found that supporting simultaneous tasks comes
with a trade-off: presenting more than two tasks reduces overall execution time, but also increases cogni-tive load.
When time-critical and simultaneous tasks are also spatially distributed and users have to switch work-stations to
perform them, location-based AR visualizations are not sufficient and support must be provided by additional
visual cues in the periphery. In this context, we have shown that existing peripheral AR cues can be en-coded with
temporal information. In addition, a high degree of human adaptability is required when time-critical, simultaneous
and spatially distributed tasks have to be taken over because the originally responsible user has to leave the workplace.
Betreuer: Prof. Dr. Susanne Boll