Human Factors Engineering

1
Human Factors Engineering

• Ken Funk
• Associate Professor of Industrial and
  Manufacturing Engineering

2
Kenneth H. Funk II, PhD

• Education
• BA, Biology, Taylor University
• MS and PhD, Industrial and Systems Engineering,
  The Ohio State University
• Teaching
• Computer applications
• Senior projects
• Human Factors Engineering
• Research
• Submarine Command Support System (US Navy)
• F/A-18 IAAS Task Support System (US Navy)
• Flight Deck Automation Issues (FAA)
• Cockpit Task Management (NASA)
• Operating Room Human-Machine Systems Engineering

3
What is Human Factors Engineering?

• Human Factors Engineering (HFE) is concerned with
  the design of human-machine systems.
• HFE seeks to achieve high levels of system
• effectiveness and
• safety.
• HFE explicitly considers human
• requirements,
• characteristics,
• capabilities,
• limitations,
• wants, and preferences.

4
(No Transcript)
5
Accidents by Primary CauseHull Loss Accidents -
Worldwide Commercial Jet Fleet 1988 Through 1997
(Source 1997 Statistical Summary of Commercial
Jet Airplane Accidents, Boeing Commercial
Airplane Group)
6
Cockpit Task Management Errors Initial
Observations

• Cockpit (flight deck) is a multitask environment
• aviate
• navigate
• communicate
• manage systems
• Results of distraction, preoccupation
• Everglades L-1011 accident
• many incidents

7
Cockpit Task Management Research

• CTM Errors in Aircraft Accidents (1991)
• 80 CTM errors in 76 (23) of 324 accidents
• CTM Errors in Critical, In-Flight Incidents
  (1993)
• 349 CTM errors in 231 (49) of 470 incident
  reports
• Part-Task Flight Simulator Study (1996)
• CTM error rate increases with workload
• ASRS Study of CTM and Automation (1998)
• Task prioritization error rate higher in advanced
  technology reports
• Findings
• CTM is a significant factor in flight safety
• CTM can potentially be improved

8
Statement of Need

• Computer-Based CTM aid is needed to
• maintain a current model of aircraft state and
  current cockpit tasks,
• monitor task state and status,
• compute task priority,
• remind the flightcrew of all tasks that should be
  in progress, and
• suggest that the flightcrew attend to tasks that
  do not show satisfactory progress.
• leave the pilot in control
• ? the AgendaManager (AMgr)

9
Simulator(with Engine Indication and Crew
Alerting System -- EICAS)
10
AMgr Architecture and Function
11
AMgr Display(replaced EICAS)
12
AMgr Operation

• simulator runs
• pilot declares goals via ATC acknowledgements
• System Actor Agents instantiate Goal Agents
• Goal Agents watch for goal conflicts
• Function Agents assess function status
• AgendaManager informs pilot via display

13
Test and Evaluation

• Objective compare AMgt performance (AMgr vs
  EICAS)
• Apparatus
• flight simulator
• AMgr
• Subjects 8 airline pilots
• Scenarios
• EUG to PDX
• PDX to Eugene
• Primary factor monitoring and alerting condition
• AMgr
• EICAS

14
Evaluation Results
15
Interaction Plot Pilot Performance vs. Display
Type
16
Interaction Plot Pilot Speed vs. Display Type
17
Alternative ApproachTask Management Training

• Is task prioritization trainable?
• Research suggests that voluntary control of
  attention is a trainable skill
• Objective
• Develop and evaluate a CTM training program to
  improve task prioritization performance.

18
Methodology

• Participants
• 12 General Aviation pilots, IFR rated, with at
  least 100 hrs pilot-in-command total time.
• Recruited through flyers and word of mouth
• Oregon State (Corvallis, Albany, Salem, Eugene,
  Portland)
• Apparatus Microsoft Flight Simulator 2000
• 3 monitors, Flight Yoke, Throttles, and Rudder
  Pedals
• IFR conditions
• Two flight scenarios

19
Lab Setup
20
Participant Display(C-182RG)

21
Experimenters Display

22
Experimental Groups

• Control Group No Training
• Descriptive Group CTM lecture
• Multi-tasking
• Attention
• CTM
• Task Prioritization errors
• Accident/Incident examples
• What to be aware of.
• Prescriptive Group
• CTM lecture
• APE procedure

23
APEAssess Prioritize Execute
A P E

• Let the APE help you
• Assess the situation
• aircraft systems, environment, tasks, procedures
• Whats going on? What should I be doing?
• Prioritize your tasks
• Aviate Is my aircraft in control?
• Navigate Do I know where I am and where Im
  going?
• Communicate Have I communicated or received
  important information?
• Manage systems Are my systems okay?
• Execute the high priority tasks Now.
• Invoke the APE frequently.
• Think out loud.

24
Experimental Procedure

• Initial briefing, informed consent
• Initial 30-minute simulator training
• Pre-training flight
• CTM training (break for control group)
• Additional 30-minute simulator training
• Post-training flight (different scenario)
• Post-experiment questionnaire

25
Primary Dependent Measure

• Task prioritization error rate
• 19 Task prioritization challenges, e.g.
• clearance near end of climb
• bust altitude? (/- 200 ft)

26
Data Collection

• Flight Data Recorder
• Videotape
• Observation
• Data reduction to
• task prioritization error rate
• prospective memory recall rate

27
Interaction Plot(task prioritization error rate)
28
Possible Interpretations

• Results may have two interpretations
• CTM training did improve task prioritization
  performance.
• CTM training did not improve task prioritization.
• Floor effect
• MSFS experience
• Age
• Research favors first interpretation
• ANOVA results
• t-tests
• Potential for better control group performance
  was there.
• Additional tests

29
Follow-Up Experiment

• Four groups
• Control 1 (no training)
• Control 2 (no training, but knowledge of measure)
• Descriptive
• Prescriptive (slightly modified APE)
• More participants (20 5 per group)
• Pre-experiment test
• short MSFS flight, pre-experiment questionnaire
• assign so as to yield equivalent groups
• Slightly shorter scenarios