|
Decision
support systems for control
rooms combats information-overload
effects of escalating levels
of automation
At
Urban Transport 2004 (Dresden
19-21 May), Cambridge Consultants
Ltd (CCL) will demonstrate the
role of decision support systems
for optimising the efficiency
of real-time control rooms.
Using
a model developed in conjunction
with a major European metro
operator, CCL's agent software
demonstrator shows how intelligent
tools help controllers to combat
the real-life stresses of centralised
control strategies, and to remain
focused on strategic issues.
As the
level of real-time automation
increases, and centralised control
rooms in areas such as processing
plants or mass transit networks
become linked to more and more
sensors and systems, it becomes
easy for the typical exception-handling
workload to be swamped by events.
A common result is that the
attention of supervisory staff
can be diverted from core or
critical operations.
"If it's not implemented
carefully, automation can funnel
unworkable quantities of data
and decisions into a centralised
control room, where it becomes
easy for operators lose sight
of the goals," notes Jon
Garnsworthy, Head of Transport
Systems at CCL. "We expect
decision support systems to
become increasingly relevant
for automation suppliers and
end-users, and we anticipate
rapid growth for the technology
in areas including mass transit,
plant control, utilities, and
defence."
CCL's
decision support system (DSS)
demonstrator, which runs on
a Windows platform, uses the
JACK Agent language system to
provide interactive decision
support for the control of a
city metro network. This prototype
application interactively assists
controllers on the detailed
implementation of their chosen
recovery strategy following
an incident. By presenting operators
with relevant and timely advice
and options, the DSS has the
potential to allow less skilled
operators to reach similar conclusions
to the most experienced operators,
in the same timescales. In the
mass transit environment, the
support provided by CCL's DSS
demonstrator embodies experience
of a wide range of emergency
conditions ranging from bomb
scares to train breakdowns,
with their enormous implications
ranging from passenger safety
to revenue and customer satisfaction.
A forward simulation tool additionally
helps the operators to anticipate
short and longer-term performance
issues and propose refinements.
Each agent has an internal structure
based on Beliefs, Desires and
Intentions. CCL's overall decision
support system for metro management
has three layers of functionality:
the user presentation layer
which is targeted toward decision-making
activities; the service layer
which helps operators to determine
the overall strategies that
should be considered; and a
layer which interfaces with
railway resources including
crew and train scheduling, and
station announcements.
|