sleep/wake research centre moe tika/moe pai philippa gander sleep/wake research centre, massey...
DESCRIPTION
SLEEP/WAKE Research Centre MOE TIKA/MOE PAI What to Measure in FRMS operational data / crew fatigue measures Operators need to be able to: Identify (monitor) level of fatigue hazard(s)Identify (monitor) level of fatigue hazard(s) Assess fatigue-related safety riskAssess fatigue-related safety risk Probability/severityProbability/severity ICAO tolerability matrixICAO tolerability matrix Define and monitor fatigue safety performance indicatorsDefine and monitor fatigue safety performance indicators Investigate role of fatigue in incidents/accidentsInvestigate role of fatigue in incidents/accidentsTRANSCRIPT
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Philippa GanderSleep/Wake Research Centre, Massey University
Jim MangieDelta Air Lines
SCIENCE IN AN OPERATIONAL SETTING
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Outline
• What to measure
• When to measure
• Interpreting the data• Safety Performance Indicators
• Conclusions
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
What to Measure in FRMSoperational data / crew fatigue
measuresOperators need to be able to:
• Identify (monitor) level of fatigue hazard(s)
• Assess fatigue-related safety risk• Probability/severity• ICAO tolerability matrix
• Define and monitor fatigue safety performance indicators
• Investigate role of fatigue in incidents/accidents
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
What to Measure: Crew Fatigue
ICAO definition of fatigue:A physiological state of reduced mental or physical performance capability resulting from sleep loss or extended wakefulness, circadian phase, or workload (mental and/or physical activity) that can impair a crew member’s alertness and ability to safely operate an aircraft or perform safety related duties.
Functional status (subjective ,objective) Sleep history Circadian phase Workload
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Samn-Perelli Fatigue Ratings
1=fully alert, wide awake
2=very lively, responsive, but not at peak
3=OK, somewhat refreshed
4=a little tired, less than fresh
5=moderately tired, let down
6=extremely tired, difficult to concentrate
7=completely exhausted, unable to function effectively
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Karolinska Sleepiness Ratings
1= extremely alert
2=
3= alert
4=
5= neither sleepy nor alert
6=
7= sleepy, but no difficulty remaining awake
8=
9= extremely sleepy, fighting sleep
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
PVT Performance (Reaction Speed)
Press this if you are right-handed
Press this if you are left-handed
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Sleep: Actigraphy and Diaries
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
When to Measure Crew Fatigue• Monitoring for FRM Processes
• In response to a series of fatigue reports• to identify extent and severity of hazard
• In response to an incident• SPIs to evaluate effectiveness of fatigue mitigations and
controls• Validation of a new route
• Monitoring for FRMS Assurance Processes• SPIs set in FRMS Policy objectives• SPIs for regulatory audit
Crew fatigue measures may not always be needed
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Safety Performance Indicators(SPIs)
5 FRMS Components
Fatigue Safety
Action Group
3. Risk Management Processes
1. Policy
4.Safety Assurance Processes
5. Promotion Processes
2. Documentation
SMS
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Comparing fatigue on ULR and LR flights
Safety Performance Indicators: Example
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
TLV-ATL
LOS-ATL
LR inbound
JNB-A
TL
DXB-ATL
ATL-DXB
ATL-JNB
LR outbound
ATL-LOS
SIN-LAX
ATL-TLV
LAX-SIN
0
2
4
6
8
10
12
LandingRelief
Median sleep (hrs) 24 hrs prior to duty
Fatigue Status at Start of DutySPI – sleep in the last 24 hrs
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
TLV-ATL
LR inbound
JNB-A
TL
DXB-ATL
LOS-ATL
ATL-JNB
SIN-LAX
ATL-TLV
LAX-SIN
ATL-DXB
LR outbound
ATL-LOS2
2.5
3
3.5
4
4.5
5
LandingRelief
mean PVT reac-tion speed
Fatigue Status at Start of DutySPI – mean PVT reaction speed early in flight
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
ATL-JNB
LAX-SIN
ATL-DXB
JNB-A
TL
TLV-ATL
DXB-ATL
LR inbound
LR outbound
ATL-TLV
LOS-ATL
ATL-LOS
SIN-LAX
00.5
11.5
22.5
33.5
44.5
LandingRelief
Median sleep in flight (hrs)
Fatigue Status at Top of ClimbSPI – median total in-flight sleep
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
ATL-JNB
DXB-ATL
ATL-DXB
ATL-TLV
JNB-A
TL
LR outbound
LAX-SIN
ATL-LOS
LR inbound
SIN-LAX
LOS-ATL
TLV-ATL
2
2.5
3
3.5
4
4.5
5
LandingRelief
mean PVT reac-tion speed
Fatigue Status at Top of ClimbSPI – mean PVT reaction speed late in flight
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
ATL-TLV
TLV-ATL
ATL-JNB
JNB-A
TL
LR inbound
LOS-ATL
LAX-SIN
SIN-LAX
DXB-ATL
LR outbound
ATL-LOS
ATL-DXB
05
101520253035404550
LandingRelief
% crew-members
Fatigue Status at Top of ClimbSPI – % crew with KSS ≥ 7
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Operational SPIs: Examples
• Track data on number of :• exceedances of planned crew duty day (e.g. > 14 hrs)• flight duty periods ending > 30 mins later than scheduled• flight duty periods starting / ending within window of
circadian low (WOCL)• reserve crew call-outs (on particular flights, at a particular
crew base, etc)
Monitoring of FRM processesMonitoring of FRMS safety assurance processes
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Conclusions• Challenge
• Developing in-house expertise versus using consultants ($$)
• Data sharing• Operators don’t have to reinvent the wheel• Rich data source for improving fatigue science
• Needs • Paradigms for data sharing• Better indicators of fatigue-related performance impairment• Better fatigue measurement technologies• Better fatigue risk assessment (safety consequences of
being fatigued in a given context)• Better fatigue risk controls and mitigations
Cooperation is the key
SLEEP/WAKE Research CentreMOE TIKA/MOE PAI
Acknowledgements
• Dr Jarnail Singh, Singapore CAA
• Participants in B-777 study
• Scientific Steering Committee, FAA 3-airline study