airspace and traffic samples march 20, 2002 · overview • (my soap-box) – general approach –...
TRANSCRIPT
Airspace and Traffic Samples
March 20, 2002
Ben WillemsEngineering Research Psychologist
NAS Human Factors GroupFAA William J. Hughes Technical Center
Overview• (my soap-box)
– General Approach– Measures
• Generic Airspace• Levels of Traffic• Scripts• Traffic Samples• Learning Effects
General Approach
To maximize our gain:
• Repeated Measures Design• Many experiments folded into one.
– Several aspects of controller behavior and performance
– Analyses of different aspects independently to form a “profile”
General Approach (continued)
Measures• Questionnaires and Self-Ratings• Subject Matter Expert Observer Ratings• Workload• Audio-Video Recordings• Communications• Standard Simulation (System) Measures• Efficiency• Situational Awareness• Visual Scanning
Fidelity
• How real for realism?• Operational systems or something
with the look and feel of the real thing?
• Existing airspace or something with the look and feel of real airspace?
DSAR1 ImplementationDSAR TGF/HOST/DSR/URET
Peripheral Adapter Module
Replacement Item
HOST
HOSTInterfaceDevice
UserRequest
EvaluationTool
URET Controller Station
URET Supervisor Station
LANInterface
Unit
DSRController
Station
DSRController
StationDSRTokenring
Sniffer
Middleware
DESIREE
NAS
URET
SAVANT
DSR
TargetGeneration
Facility
SimulationPilot Station
ExerciseControl
XPVD
RadarBox
RadarBox
RadarBox
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
TGF
SABET Implementation
DESIREETGF
TargetGeneration
Facility
SimulationPilot Station Exercise
Control
XPVD
DSRController
Station
DSRController
Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
SimulationPilot Station
DSRController
Station
DSRController
Station
DESIREE
Generic Airspace
• Not a new concept– Aerocenter High as used at the FAA
Academy– ZCY or Universal Data Set (UDS) for
Operational Test and Evaluation (documentation dated 1973)
– Initiated in our laboratory as separate enroute and terminal airspace in 1994/95
Generic Airspace: Advantages
• Easy to learn• Controllers from any airspace• Controllers start at level playing field• Results generalize to NAS
Generic Airspace: Disadvantages
• All participants have to learn• Specific airspace related habits may
not be transferred to the generic airspace
• No airspace specific results
Generic Airspace: When to Use
• Concept Research– Change in systems– Change in procedures
• Do not use it when you are about to implement a facility specific system procedure
Generic Enroute Sector
• Used with ZJX controllers– tested on own airspace and generic
airspace to determine if using generic airspace affected controller behavior
– to determine time needed to train on generic airspace
• Naming convention based on compass rose
Generic Enroute Sector
Atlantic City TRACON
Generic TRACON Airspace
• Four Corner Post Layout• Naming convention based on
compass rose
Generic TRACON Airspace
What Happened Next...
• Go study automation– User Request Evaluation Tool (URET)
• Not connected to our simulator• Available with Display System
Replacement (DSR) • How to get a good sample size of non-
URET-trained controllers
• Create a Generic Center
Generic Airspace for DSAR1
GEND2
GEN
IOW INDIN COLUM OHO
BOSTN
MAS
DETRO
MIC
M INSP
MIN
NEB
LINCO
MONIDA
NEV
NMX
ARK
PEN
BWINT
MARLOUIV
KTYWVAILL
BUTTE
DESMN
CHIGO WHEEL
J 26
J 22
J 25
J 12
J 1
J 15
J 23
J 24
J 21
J 13
J 1
J 23
J 26J 13
GENERACENTER (ZGN)
1/10/00
KANCY
Southwest High
FL240 & above
127.0 / 327.0
Northwest High
FL240 & above
128.0 / 328.0
Northeast High FL240 & ABV
124.0 / 324.0
Southeast High FL240 & above
125.0 / 325.0
Genera Low
160 -FL230
126.0 / 326.0
North High
FL240 & above
123.0 / 323.0
X
X
X
X
X
X
X
X
X
Controller chart - not drawn to scale
DSAR1 Implementation (continued)
J22
J1
J12
J12J30
J15
MID
DWN
J13
J1UPT
NTH
OZARK
BAITSR1002
R1003
R1005
R1001
OZARKR1006
DROPZONE
GEN
ILL
GEND2
WVA
KANCY
CHIGO
NWNWT
WHEEL
EST
SGFMAIT
LAKES
PONDS
Genera Genera LowLow
High126.0
En Route127.0
Approach 125.0
IND
Generic Airspace for DARC Study
Generic Airspace for SABET
02
J23
J100
J12
J15
J110
J10J120
J70
J60 J15
J23
J13
J21
J22
J30
JAX
KTY
J21
J70
WVA
MAR
NEV
NMX
CAL
ARK
BUTTE
J26
J26
MIC
KANCY
FLA
J15
27
28
OHO
MINSP
MONWAS
3303
30
05
04
14
07
GEN
CRW
ESTWST
IOW
GEND2
BOSTN
06
31
J50
J11
STH
ILL
LINCO
J50
J100
J120
120.06
120.03
120.04
120.30
120.28
120.33120.02
120.03
120.14
120.27
120.07
120.05DODGE
CIN
SPL
J11
08
22
16
J23
J23 J23
J100
J15
J10J120
J70
J60J15
J13
J21
J22
J25
J70
CAL
J26
J26
FLA
J30
TEX
MIS
JAX
LIT
GEND2
ILL
GEN
SPL
J110KTY
WVA
MAR
BOSTN
IDA
NEV
NMX ARK
BUTTE
MIC
KANCY
NJY
OHO
VIR
TEN
ORE
MAI
MINSP
MIN
MONWAS
EST
WST
IOW
CRW
OHN
MAS
LINCOPEN
STH
NTH
WHEEL
J13
J1
J1J11 J11
J20
J24
J15
SET J50
J10
GEO
J12
J100
CAR
J21
J20
Genera Center3-Sector Configuration
02J23
J23
J100
J12
J15
J110
J10J120
J70
J60 J15
J23
J13
J21
J22
J30
JAX
KTY
J21
J70
WVA
MAR
IDA
NEV
NMX
CAL
ARK
BUTTE
J26
J26
MIC
KANCY
NJY
FLA
J15
27
28
OHO
VIR
TEN
TEX
MIS
ORE
MAI
MINSP
MINMON
WAS
3303
30
05
04
1429
07
GEN
CRW
ESTWST
IOW
GEND2
BOSTN
MAS
06
31
J10
J50
J24
J20
J11 J11
J25
GEO
STH
ILL
LINCO
J50
J100
J120
Genera Center2-Sector Configuration
Levels of Traffic
• We use an “operational” definition, a.k.a. “ask the experts”
• Personally I prefer to train controllers on “moderate” traffic and use levels that are “low” and “high” for experimental scenarios
• It would be a lot better if we could better define/characterize our scenarios in a more objective way
Scripts
• ...or stuff happens• when using scripts, bring in some
“naive” controllers in a pilot study to see if the scripts actually work
• make sure that you have your measures in place that capture the essence of the script
Traffic Samples• Real airspace
– Sampling Strategy– Data Collection– Data Screening– Scenario Generation– Scenario Shakedown
• Generic airspace– Define Traffic Requirements– GYOT (generate your own traffic)– Scenario Shakedown
Learning Effects• Control for order effects
– counter balancing– rotation of scenarios under experimental conditions
• Sometimes (in my opinion often) you cannot get away with using multiple scenarios to prevent learning effects:– if you are interested in local events or situations, e.g.:
• with this enhancement, are aircraft better spaced over a fix?• With change in procedures, will aircraft fly more efficiently?