Text Size: - Smaller | + Larger
Forum
Back To Home Page
Previous Research
This is an extract from Hina Keval's research:
"
Despite the increased usage of surveillance systems and the technological
advances, there is
currently no conceptual basis and little evidence to assess how well CCTV
actually works for the
purposes for which it is deployed. It is important to identify whether CCTV
systems and applications meet stakeholder goals, and support human operators effectively,
in attaining the goals for which the system is set up. This exploratory paper highlights the
early findings found in CCTV control rooms. A series of ethnographic observations along with
semistructured interviews were carried out at six CCTV public surveillance control rooms -
six managers, six supervisors and 25 operators were interviewed. Findings reveal that current
control room systems are not designed to support operator and system communication and
collaboration when performing their tasks. Poor linked technology and a lack of task
coordination was evident between operators and external agencies such as police operators and local businesses. Several
Human Computer Interaction (HCI) issues were uncovered from the fieldwork. Findings will be
used to form a set of best-practice CCTV control room design recommendations."
ForumBack To Home Page
Previous Research
Top Of Page
1 Background
1.1 Research Problem
A number of studies have investigated how dynamic systems and processes are
managed by
operators in control room environments such as air traffic control and nuclear
power plant
control centres. Despite this, there appears to be very little HCI research
in security and
surveillance control rooms. In the last decade, we have seen several changes
take place in
security - with a rise in crime rates, and the type and severity of crime
events have also changed.
Consequently the public's perception and fear towards crime have also changed.
More funding
is available for CCTV and more advanced CCTV technology is also available.
These
changes
have led the research discussed in this paper to form a number of important
questions: How are
CCTV control room managers managing new digital technologies? Do control room
operators
understand how to use digital technology and multiple systems/tools? The
overall question this
research attempts to tackle through exploratory cognitive ethnography is:
Are public surveillance
control rooms operating effectively and efficiently. This research focuses
on the difficulties
operator face with control and co-ordination of surveillance tasks. The
relationship and
performance of communication and collaboration between operators and
external agencies,
information management and task performance effectiveness and efficiency
were explored using
cognitive ethnography techniques (see Hollan, Hutchins and Kirsh., 1999).
Forum
Back To
Home Page
Previous Research
Top Of
Page
1.2 Previous Control Room Research
Luff and Heath (1999) examined how control room operators used CCTV and
other technologies
within an underground transport control room environment. Luff found that
the control room
technology was difficult to manage because there were "so many separate
interconnected
systems and the use of these systems are thoroughly embedded within the
many disparate
activities of the personnel." From this study, various user-system
interactions were considered,
however very little focus was placed on the HCI barriers to task operations.
Gill (2005)
attempted to assess the impact of public CCTV surveillance as a whole on
crime and touched on
the technology issues in control rooms. Gill reported that many control
rooms had street cameras
positioned in poor locations, video recording rates were too low and
communication flow
between operators and other CCTV stakeholders was very low. The research
focussed on
whether CCTV as a whole effectively reduced crime and not the effectiveness
of control room
operator interactions where technology is heavily used.
Currently, we know very little about how end-users utilise modern CCTV
technologies to
perform their tasks within busy CCTV control rooms. The focus of the research
presented in this
paper looks at the effectiveness of CCTV technology within several city
centre control rooms.
Ethnography was selected as a research method for assessing the effectiveness
and efficiency of
control room task operations for two main reasons: (1) Many Computer
Supported Cooperative
Work (CSCW) systems often fail because the design process excludes the
social work context,
HCI and human factors issues; (2) "Many of the existing methods fail to
sufficiently recognise
the social setting of the social nature of work," (Hughes et al., 1993) -
i.e. task analysis and
interviewing.
Forum
Back To
Home Page
Previous Research
Top Of
Page
The nature of the observations adopted was something Hughes
termed as "quick
and dirty ethnography." This method was considered appropriate as the
current research
contributions of this research (the best-practice design recommendations)
aims to inform control
room managers broader issues related to the acceptability and usability of
CCTV control room
design. The HCI aspect of this work was studied from a theoretical perspective
of distribution
cognition (see Hollan et al., 1999). This perspective works on the basis which
"seeks to
understand the organisation of cognitive systems." Cognitive systems can be
distributed across
members of a social group (CCTV stakeholders: i.e. operators, police,
management, public etc.)
and involve the co-ordination between internal and external (material or
environmental)
structures.
2 Communication and Collaborative Tasks: Field Observations at Six Control
Rooms
The aim of the fieldwork was to obtain a better understanding of the
human-to-human and
human-to-system communication, coordination, and control mechanisms in CCTV
control rooms
when information flow is high. The control rooms visited were set up to
support several
stakeholders such as the local police staff and the community i.e. clubs,
pubs, shops and the
general public. The coordination of reactive and proactive surveillance
tasks such as patrolling
CCTV screens, responding to police radio calls, and sharing imagery with
police when needed
were achieved in all of the control rooms visited using the same type of
equipment.
Forum
Back To
Home Page
Previous Research
Top Of
Page
Equipment
included: CCTV wall monitors; camera controllers; a Personal computer
(PC); 'spot' monitors
(these are 2-5 video monitors located directly in front of operators which
can be used to grab a
CCTV video output selected from the wall monitors); radio; and telephones.
Despite the
similarities, it was found that the workstation layout, equipment set-up,
communication flow
between operators and external agencies and the design of the tasks differed
between control
rooms. A series of overt observations and semi-structured interviews with a
total of 25 CCTV
operators, six managers and six security supervisors were made by one field
observer. Five of
the control rooms were based in London and one outside London. Observations
took place per
control room on average over a period of five hours. Visits were made to
control rooms during
morning, afternoon, and evening shifts. This was so that the operator's
tasks and activities could
be analysed under different situations. For example, different cameras
were used under different
lighting conditions and different criminal and suspicious activities were
observed at different
times. An observation checklist of ideas and areas of interest to the
observer was followed (tasks,
equipment, communication, workspaces etc.) which was used as structured
protocol for the
observation exercise. Operators and supervisors throughout the observation
period were
informally asked questions about 'what was going and why they did tasks in
that way.'
Responses were recorded and supported the observation notes.
2.1 Technology and Setup: Mapping Geographical Information
A recurring problem found in a majority of the control rooms was the way in
which operators'
located CCTV screen(s) when attempting to follow a vehicle or person.
Operators used paperbased
geographical street maps with lists indicating the street names and camera
numbers.
Forum
Back To
Home Page
Previous Research
Top Of
Page
Many
of the operators said for this type of task, "having good local area
knowledge was important."
Despite this, a minority of the operators lived in the areas they observed
on-screen. Operators
regularly shouted to colleagues across the room if they were stuck and could
not recall the
camera number or its location. Operators shouted louder and in a panic-like
tone particularly
when communicating with police operators via telephone or radio to follow
targets of interest onscreen.
The use of physical paper maps is a risky, ineffective, and inefficient
method for
searching and tracking targets. Paper maps can go astray and losing a map
would lead to guess
work, which added unnecessary time to the task. Several managers reported
that considerable
funding was granted by senior councillors and the Home Office for adding
additional CCTV
cameras to their systems. Many of the operators complained that there were
"too many cameras
to cope with" and found these additional cameras were often not updated onto
the paper maps
and camera lists.
A method to improve the efficiency and effectiveness of searching and
selecting cameras
would be to link the camera monitor views with a graphical user interface
(GUI) linked to a
geographical map of the surveillance areas linked to a comprehensive
database of camera names
and street locations. Such a method would avoid operators mishearing
numbers and locations
when shouting information across the control room; it would also avoid the
risk of confusion. A
simple coordinated tool can be used to allow operators to communicate with
a common
understanding of the situation.
2.2 Reactive Surveillance: Information Overload and Inconsistent Radio
Language
Forum
Back To
Home Page
Previous Research
Top Of
Page
Often, control rooms are thought of as small, dark underground rooms filled
with surveillance
cameras wall to wall, with a handful of operators idly waiting for something
to happen on video
monitors. This is untrue. In fact, tasks are not so much video driven and are
more audio driven.
Operators perform two key surveillance tasks - proactive surveillance
(watching and waiting for
something to happen) and reactive surveillance (responding to alerts from
outside control room
to react to a crime or suspicious event). From the two, the most frequent
surveillance task
operators perform was the reactive surveillance task. The most heavily used
communication tool
used for this task was the police and business radio. Close observations of
operator actions and
operator remarks showed that there were clear signs of cognitive overload
with radios and
telephones. Several operators commented that "the control room radio has too
many different
channels assigned and sometimes it can be too confusing what's going on,
especially when the
phone is going off as well." Operators also complained that business radio
users such as city
centre shop managers gave too much unnecessary information and that they
did not give clear
descriptions of targets. Excessive radio groups and poor information flow
between users are two
of the most common causes of cognitive overload: (1) too much information
supply and (2) too
much information demand (Kirsh, 2001). A simple solution to minimise audio
information
overload would be to distribute radio channels across the numbers operators
and the level of
activity over the day proportionally. Distinguishable audio tones is useful
method for funnelling
initial incoming contact that is made with the control room, so that
operators can prioritise and
respond to radio calls accordingly. Feedback and training should also be
provided to all radio
users, so that consistent language is used such as phonetics and identity
codes which are
commonly used by UK police forces
Forum
Back To
Home Page
Previous Research
Top Of
Page
Conclusions
The field work described here highlights how computerised technology in city
centre CCTV
public surveillance control rooms are set-up and designed without considering
what tasks are
performed by operators. "Any design of computer systems for control room,
which does not
support intrateam communication and coordination, is very likely to fail in
the long run,"
(Garbis, 2000). Ineffective workplace designs currently in modern control
rooms where
information communication technology (ICT) is heavily used, can affect both
internal and
external communication - affecting the control and coordination of tasks.
A combination of
system re-design where existing systems can be linked together and training
across stakeholders
concerning the system can improve task effectiveness, efficiency, and overall
work performance.
Further control room visits are in progress. Findings will be used to form
a set of best-practice
recommendations for CCTV control room management."
[Hina Keval University College London]
Most Recent And Up-to-date
October 2007 information.
Reproduced By Kind Permission Of
Privacy International Web Site:
Numerous criminological studies, including those commissioned by the Home Office, the interior ministry responsible for policing, and by independent groups such as the National Association for the Care and Resettlement of Offenders (NACRO) have consistently concluded that the impact of surveillance cameras is marginal and largely illusory.
Their chief benefit, according to researchers, is in deterring some types of low level "opportunistic" crime, a benefit that could be more inexpensively and effectively achieved through other means.
The most recent of these reports, the October 2007 study on the national CCTV strategy concludes that eight images out of ten supplied to the police from closed-circuit television do not help to identify criminals. The report, compiled by the Home Office and the Association of Chief Police Officers, said:
"Anecdotal evidence suggests that over 80 per cent of the CCTV footage
supplied to the police is far from ideal, especially if it is being used
for primary identification or [where] identities are unknown and
identification is being sought.[1]
More »
[1] National CCTV Strategy, ACPO and the Home Office, October 2007, page 12.
ForumBack To Home Page
Previous Research
Top Of Page
Bibliography:
Garbis, C. 2000. “Communication and coordination through public and private representations in control rooms,” in CHI '00 Extended Abstracts on Human Factors in Computing Systems (The Hague, The Netherlands, April 01 - 06, 2000), ACM Press, New York, NY, pp 67-68, 2000.
Gill, M., Allen, A., Jessiman, J., Swain, D., Hemming, M., Kara, D and Little, R. Methods in assessing the impact of CCTV, Home Office report, no 17, 2005.
Hollan, J., Hutchins, E. & Kirsh, D. “Distributed Cognition: Toward a New Foundation for Human Computer Interaction Research,” in ACM Trans. on Computer-Human Interaction, vol. 7, no.2, pp. 174-196, 2000.
Hughes, J., A, King, V., Rodden, T. and Andersen, H. “Moving out from the Control Room: ethnography in system design,” in proceedings of CSCW '94, Chapel Hill, pp. 429-439, 1994.
Kirsh, D. "A Few Thoughts on Cognitive Overload," Intellectia, vol. 1, no. 30, pp. 19-51, 2000.
Luff, P and Heath, C. "Surveying the Scene: The Monitoring Practices of Staff Control Rooms,” in proceedings of People in Control: An International Conference on Human Interfaces in Control Rooms, Cockpits and Command Centres, (Eds), IEE Press, 1999.
1. Long-term Trends in Violence against the Person:
2. Analysis of costs and benefits guidance for evaluators:
3. Measuring inputs guidance for evaluators:
4. www.neighbourhood.statistics.gov.uk
5. www.homeoffice.gov.uk/rds/recordedcrime1.html
