Human Factors and Aviation Security

Sandra G. Hart
NASA-Ames Research Center
Moffett Field, CA

SG Hart (12/1/01) 1
Overview Human factors is a common element across
past failures and future solutions

 Magnitude of the problem

 Underlying differences between safety and
security
 Use of human factors tools to identify
vulnerabilities
 Ensure that the humans responsible for each
layer of defenses against future threats are
part of the solution, not the problem
 Baggage screeners
 Monitoring/maintaining secure areas
 Air/ground and cockpit/cabin interactions
 Assess the human and system impact of
technology and procedural solutions
 Human Factors in Aviation Security Act

SG Hart (12/1/01) 2
US Aviation at Worldwide, there are more than 800 airlines
employing more than 150,000 pilots and 16,000
the Turn of airplanes, flying into more than 1,350 major
airports in more than 200 countries
the Century
Safety
 1 hull loss/1 million P-121 departures

Capacity
 14,000 a/c in US civil fleet
 2M airline passengers/day
 77M tons of cargo carried each day
 60,000 GA flights/day
 4000-5000 aircraft aloft at any time
 66% flights concentrated at 31 hub
airports

Reliability
 1 flight in 4 delayed

Infrastructure
 ~325 air traffic control facilities
SG Hart (12/1/01) 3
 Baggage screeners
Aviation • Numbers: There are ~30,000 baggage screeners, many of
whom are not citizens (60-80% in some major airports) and
Security at all of whom are underpaid
the Turn of • Training: Baggage screeners receive as little as 12 hr of
classroom training
the Century • Tenure: Turnover at 19 largest airports averages 126%/year
(the worst situation was at St Louis at 425%)

Baggage screening equipment

• 47 out of 420 airports have explosive detection systems
• It will cost $2B to fully equip airports with CT scanners
• Currently, <10% checked bags screened

Baggage screener performance

• Screeners detect about 2000 firearms each year
• In GAO test, 20% of the test weapons were not detected

Access to secure areas

• 1M people have access to airport secure areas (~ 9%
badges are in wrong hands)
• 750, 000 airport & airline personnel will require back- ground
checks, access to secure areas
• In DoT test, security was breached 117 tries out of 173 (68%)

SG Hart (12/1/01) 4
 Safety vs Security:
Parallel, Independent Redundancies Ensure Safety
Pre-flight Wx forecast Automation
Inflight Wx/PiReps Redundant hardware
Wx radar, Pilots’ eyes Redundant software
GPWS, MSAW, TCAS Manual reversion modes

Instrument scan Mutual Monitoring

Alerts & warnings

Flight plans
Mutual Monitoring

Procedures
Checklists
Crosscheck Back-up systems
Training Alerts & warnings
CRM Readbacks
SoPs

… against threats to safety that are neither deliberate

nor the result of intelligence and planning
SG Hart (12/1/01) 5
 Safety vs Security:
The security “net” is vulnerable because lacks redundancy

Threat Vectors

Nodes in the Security Net

… against threats to security that are ill-defined, evolving and

the result of deliberate and intentional actions
SG Hart (12/1/01) 6
 Threat Introduction Vectors

Transmitted information
Verbal communications
Other aircraft

Paying pax
Non-revenue pax
Carry-on luggage
Checked baggage
Unaccompanied cargo
Security personnel
Maintainers
Gate personnel Airways facilities personnel
Cleaning crews Air traffic managers
Food service Security personnel
Flight attendants Suppliers
Pilots

SG Hart (12/1/01) 7
 Specific Human Factors Targets
Analyze risks, vulnerabilities
In existing system Predict system impact
Voluntary reporting of proposed changes
system

National aviation Traffic flow baseline, flight-

Data compression algorithms path conformance monitoring
security survey
to transmit airborne/ground anomaly detection
surveillance video
ATM lessons Pros, cons of transferring a/c
Consistent, integrated control to ground, automation
learned from 9/11
air/ground & cockpit/cabin
procedures, training

Model, simulate, Alertness monitoring,

Transmitted information
evaluate impact of shift-scheduler for Paying pax Verbal communications
technologies and security personnel Non-revenue pax Other aircraft
Carry-on luggage
procedures on Checked baggage
humans in system Support formation of Unaccompanied cargo
effective security teams Maintainers
Security personnel
Positive ID to enter Gate personnel
Task design, aiding, Cleaning crews
any secure areas
training to improve Food service
check-point screening Flight attendants Airways facilities personnel
Pilots Air traffic managers
Data base mining to Security personnel
identify threat Machine vision systems Suppliers
patterns, profiles to screen baggage; alert
humans
Decision support system for
Support formation of a Detection algorithms/alerts/user
ticket agents to ID potential
“security culture” interface for baggage
security risks
screening

Visualization tools to convey important, time critical info,

& depict alternative courses of action

SG Hart (12/1/01) 8
False Scope: Infuse the rush toward technology
Promises of acquisition with a bit of (human factors)
common sense. Mitigate against unintended
Technology consequences and rigid, single-point barriers

Potential Applications:
 Independent, redundant layers of security at
each node
 Coordinate between sequential nodes in
system - - growth of evidence
 Be careful of wasting time, money and public
confidence with high-cost, high-tech
solutions

SG Hart (12/1/01) 9
System Scope: Develop system-wide baseline and trend
information to identify gaps and vulnerabilities in
Monitoring & the security system
Information
Potential Applications:
Sharing  Acquire, manage & analyze traffic flow data to
develop baselines, track system performance
 Security-related incident reporting system,
surveys
 Analyze risk and vulnerabilities in the current
aviation system
 Identify patterns and trends in passenger profiles
 Improve air and ground detection of flight-path
deviations using traffic displays
 Adapt visualization tools to convey key security
information clearly and unambiguously
 Scrutinize the safe and successful diversion and
landing of thousands of aircraft for lessons
learned and implications for the evolution
of the air traffic system

SG Hart (12/1/01) 10
Apply Scope: Apply fundamental knowledge of human
vision, cognition, attention to enhance aviation
Fundamental security systems
Knowledge
Potential Applications:
 Apply vision science to automated pattern
recognition of threats, face recognition
algorithms to alert human monitors to potential
threats, biometric identification of personnel to
control access to restricted areas
 Video-data-compression techniques to support
real-time analysis of down-linked cockpit/cabin
images
 Mitigate the pitfalls of sustained attention,
visual search and monitoring tasks

SG Hart (12/1/01) 11
Performance Scope: The field of Human Factors is based upon its
expertise in assessing human and system
Evaluation performance in simulated and operational
and Support environments. Use this expertise to assess the
impact of security solutions.

Potential Applications:
 Impact of changes in aircraft cockpit/cabin
configuration on crew coordination
 Pros and cons of proposals to remove aircraft
control authority from the flight deck
 Real-time monitoring of alertness of personnel
responsible for passenger and baggage
screening, security cameras, etc
 Air traffic control response to perceived hijacking
 Improving the quality of simulated threats used
for training and testing baggage screeners (TIP)

SG Hart (12/1/01) 12
Task Analysis, Scope: Human Factors expertise in task analysis,
Modeling & modeling, and simulation offer insights into the
skills required to perform crucial tasks, identify
Simulation functions that are candidates for automation,
and predict the human and system impacts of
conceptual and prototype designs

Potential Applications:
 Selection of more effective screeners
 Identify technologies to support inspection and
screening functions for which humans are
particularly ill-suited
 Computational & man-in-the-loop analyses of
new technologies, procedures
 Computational modeling tools to enable
incorporation of human factors in design of
security technologies and procedures

SG Hart (12/1/01) 13
Procedure Scope: Assess the impact of new regulations and
& Policy counter- terrorism philosophies on aircrew
procedures, controller procedures, and
Development air/ground interactions. Effectively integrate
new technologies and changes in procedures
into flight and ground operations.

Potential Applications:
 Consistent and coordinated flight and cabin
crew threat response procedures using existing
and future technologies
 Consistent and coordinated air/ground
responses to threats
 Develop a formal model of the security system,
modeling characteristics of human and machine
detectors to identify vulnerabilities. Estimate
the impact of security measures on passenger
throughput, and of changes in passenger load
on effectiveness

SG Hart (12/1/01) 14
Training Scope: Apply expertise in forming effective
teams to improve security personnel’s job
satisfaction and performance. Avoid
proliferation of ad hoc responses by pilots
and controllers in response to perceived
threats and the introduction of new
technologies with inadequate training.

Potential Applications:
 Selection of qualified personnel for
screening people, baggage, and cargo.
 Training to support new procedures and
technologies for cabin crew, flight deck
crew, and air traffic managers
 Adaptation of new training technologies,
such as virtual reality
 Incorporate security into safety culture
 Mitigate impact of fatigue on screeners

SG Hart (12/1/01) 15
 Summary of Topics
Underlying
Security-Specific Applications Focus
Science
System Analysis of risk, vulnerability
Voluntary reports of security issues System vulnerabilities, patterns
Monitoring &
& trends
Information Data mining and visualization
Passenger profiles
Sharing Post hoc analyses
Automated pattern recognition devices Baggage screening
Vision Biological identification systems Personnel ID
science
Efficient transmission of visual data Remote monitoring
Perception, Monitoring multiple auditory channels Baggage screening
Attention & Vigilance/Pattern recognition Video surveillance
Cognition Flight plan conformance monitoring Deviations in flight path
Real-time alertness monitoring Security personnel
Performance
Assess impact of procedures, technology, Aircraft crew coordination
evaluation
environment on operators Air/ground coordination
Procedures Distributed Teamwork Aircraft crew coordination
& Policy Transfer of control Air/ground coordination
Computational analysis of candidate
solutions Security personnel selection
Task Analysis
Model-based exploration of procedures Security, ATC, pilot procedures
& Modeling
Airspace operations
Incorporate human factors into design
Effective teamwork Security teams
Training Use of virtual reality, intelligent tutoring Flight crews
Reduce airport screener/monitors fatigue Controllers & supervisors

SG Hart (12/1/01) 16
Human Factors Issues in Aviation Security Act
• Flight deck door reinforcement/ Flight deck access
• Determine seriousness of occurrences
• Crew communication and coordination
• Warning devices
• Cabin surveillance cameras
• Cabin and flight deck defense/protective devices
• Psychology of terror (passenger behavior)
• Security training/Common strategy
• Cabin search procedures
• Transponders
• Aircraft defensive procedures/maneuvers
• Security workforce selection/background checks
• One level of security
• Universal access card/”smart ID cards” with biometrics
• Universal baggage screening/ID matching/photo manifest
• Trusted passengers/pre-identified “volunteers” with special skills

SG Hart (12/1/01) 17