Fault management during dynamic spacecraft flight: effects of cockpit display format and workload
Page 1
Fault management during dynamic spacecraft flight: effects of cockpit display format and workload
Fault Management during Dynamic Spacecraft Flight:
Effects of Cockpit Display Format and Workload
Valerie A. Huemer
San Jose State University Foundation
NASA Ames Research Center
Moffett Field, CA, U.S.A.
Valerie.A.Huemer@nasa.gov
Michael P. Matessa, Robert S. McCann
Human Factors Research and Technology Division
NASA Ames Research Center
Moffett Field, CA, U.S.A.
{Michael.P.Matessa, Robert.S.McCann}@nasa.gov
Abstract - A proposed redesign of the shuttle cockpit
display formats improves the correspondence between
system summary displays and crewmembers’ mental
models of systems architecture and functional mode.
We report the results of a part-task simulation that
assessed the impact of the redesigned displays on
participants’ ability to perform various steps in the
process of diagnosing and recovering from systems
malfunctions. Participants were airline pilots who
received a modest amount of training on the tasks
required of shuttle crews during nominal and off-
nominal ascents. With respect to fault management
performance, both errors of omission and commission
were reduced with the redesigned displays. Fault
management errors were further categorized within a
cognitive-stage information processing framework.
Error rates increased steadily from early to late stages
of processing, but more so for the current displays than
for the redesigned displays. We conclude that classifying
and analyzing errors made by participants with
relatively low levels of training provides a useful
methodology for assessing and evaluating human-
centered design modifications to spacecraft displays.
Keywords: Errors, omissions, commissions,
information processing model, spacecraft, space shuttle,
MEDS, CAU.
1 Introduction
The space shuttles flying today were designed and
built over 25 years ago. Despite several hardware
improvements since that time, including conversion to a
“glass cockpit” in 2000, crew workload remains high,
particularly in off-nominal situations. Cockpit
automation to support the crew in off-nominal
conditions is minimal, leaving most emergency
operations to be performed by the crew and mission
control personnel. For example, the cockpit caution and
warning system automatically annunciates out-of-limits
sensor readings that accompany systems malfunctions.
However, it is up to the crew and mission control to
jointly diagnose the root cause of the anomalous
reading(s), access the appropriate remedial actions in
paper flight data files (FDFs), perform the specified
actions, and verify their effectiveness. For many
malfunctions, these activities increase crew workload to
the point where crewmembers have very little margin to
deal with any additional problems.
Another factor contributing to crew workload is
that the cockpit displays were designed to accommodate
1970’s era limitations in electronic display technology
and onboard computing capability [1]. Consequently,
displays are data-source-oriented (rather than task-
oriented), so the crew is often required to navigate
through several displays to gather the information
needed to complete a task. The displays are often poorly
organized and highly cluttered, taking the form of
closely-spaced tables of alphanumeric data that require
considerable mental translation in order to infer the
current operational status or functional mode of the
onboard systems.
To address the shortcomings of the current shuttle
cockpit display formats (collectively known as the
“Multifunction Electronic Display System” [MEDS]),
the current display formats have been completely
redesigned as part of a recently completed Cockpit
Avionics Upgrade (CAU) project (unfortunately, due to
budget considerations, the display redesigns are not
going to be actually implemented on the shuttles before
they are retired in 2010). One of the fundamental human
factors principles guiding the redesign of the systems
summary displays (those displays that provide
information about systems status and system
functioning) was that these displays should be
“transparent” to the actual working system, so that the
operator can “see through” the displays to “what is going
on” with the underlying system [2]. The resulting
display formats, described by McCandless and McCann
[3], consolidate systems information onto single, task-
oriented systems summary displays to reduce the need
for display navigation. Many of these display formats
incorporate dynamic graphical depictions of systems
components, providing “at a glance” indications of the
operational status and functional mode of the subsystem.
The following illustration shows how these human
factors principles were applied in the upgraded cockpit.
7460-7803-9298-1/05/$20.00©2005 IEEE
2005 IEEE International Conference on Systems, Man and Cybernetics
Waikoloa, Hawaii October 10-12, 2005
Effects of Cockpit Display Format and Workload
Valerie A. Huemer
San Jose State University Foundation
NASA Ames Research Center
Moffett Field, CA, U.S.A.
Valerie.A.Huemer@nasa.gov
Michael P. Matessa, Robert S. McCann
Human Factors Research and Technology Division
NASA Ames Research Center
Moffett Field, CA, U.S.A.
{Michael.P.Matessa, Robert.S.McCann}@nasa.gov
Abstract - A proposed redesign of the shuttle cockpit
display formats improves the correspondence between
system summary displays and crewmembers’ mental
models of systems architecture and functional mode.
We report the results of a part-task simulation that
assessed the impact of the redesigned displays on
participants’ ability to perform various steps in the
process of diagnosing and recovering from systems
malfunctions. Participants were airline pilots who
received a modest amount of training on the tasks
required of shuttle crews during nominal and off-
nominal ascents. With respect to fault management
performance, both errors of omission and commission
were reduced with the redesigned displays. Fault
management errors were further categorized within a
cognitive-stage information processing framework.
Error rates increased steadily from early to late stages
of processing, but more so for the current displays than
for the redesigned displays. We conclude that classifying
and analyzing errors made by participants with
relatively low levels of training provides a useful
methodology for assessing and evaluating human-
centered design modifications to spacecraft displays.
Keywords: Errors, omissions, commissions,
information processing model, spacecraft, space shuttle,
MEDS, CAU.
1 Introduction
The space shuttles flying today were designed and
built over 25 years ago. Despite several hardware
improvements since that time, including conversion to a
“glass cockpit” in 2000, crew workload remains high,
particularly in off-nominal situations. Cockpit
automation to support the crew in off-nominal
conditions is minimal, leaving most emergency
operations to be performed by the crew and mission
control personnel. For example, the cockpit caution and
warning system automatically annunciates out-of-limits
sensor readings that accompany systems malfunctions.
However, it is up to the crew and mission control to
jointly diagnose the root cause of the anomalous
reading(s), access the appropriate remedial actions in
paper flight data files (FDFs), perform the specified
actions, and verify their effectiveness. For many
malfunctions, these activities increase crew workload to
the point where crewmembers have very little margin to
deal with any additional problems.
Another factor contributing to crew workload is
that the cockpit displays were designed to accommodate
1970’s era limitations in electronic display technology
and onboard computing capability [1]. Consequently,
displays are data-source-oriented (rather than task-
oriented), so the crew is often required to navigate
through several displays to gather the information
needed to complete a task. The displays are often poorly
organized and highly cluttered, taking the form of
closely-spaced tables of alphanumeric data that require
considerable mental translation in order to infer the
current operational status or functional mode of the
onboard systems.
To address the shortcomings of the current shuttle
cockpit display formats (collectively known as the
“Multifunction Electronic Display System” [MEDS]),
the current display formats have been completely
redesigned as part of a recently completed Cockpit
Avionics Upgrade (CAU) project (unfortunately, due to
budget considerations, the display redesigns are not
going to be actually implemented on the shuttles before
they are retired in 2010). One of the fundamental human
factors principles guiding the redesign of the systems
summary displays (those displays that provide
information about systems status and system
functioning) was that these displays should be
“transparent” to the actual working system, so that the
operator can “see through” the displays to “what is going
on” with the underlying system [2]. The resulting
display formats, described by McCandless and McCann
[3], consolidate systems information onto single, task-
oriented systems summary displays to reduce the need
for display navigation. Many of these display formats
incorporate dynamic graphical depictions of systems
components, providing “at a glance” indications of the
operational status and functional mode of the subsystem.
The following illustration shows how these human
factors principles were applied in the upgraded cockpit.
7460-7803-9298-1/05/$20.00©2005 IEEE
2005 IEEE International Conference on Systems, Man and Cybernetics
Waikoloa, Hawaii October 10-12, 2005
Sign up today - FREE
Mendeley saves you time finding and organizing research. Learn more
- All your research in one place
- Add and import papers easily
- Access it anywhere, anytime
Start using Mendeley in seconds!
Readership Statistics
3 Readers on Mendeley
by Discipline
33% Psychology
by Academic Status
33% Ph.D. Student
33% Researcher (at a non-Academic Institution)
33% Professor
by Country
67% United States
33% United Kingdom


