Monitoring Normal Operations

1
Monitoring Normal Operations
Presented by
Stéphane DEHARVENGT Head of Human Factors
programme

• A European perspective

2
Im a regulator, Im here to help !
3
A message from the JAA Human Factors Steering
Group (HFStG)
Regulators
European organisations AEA, AEI, ITF, IFA, ECA,
EAAPS, EAAP
Observers FAA-US, Boeing, Airbus
Active sub groups (including practitioners in
the field) Certification Design Ops
Licensing Maintenance
4
General comments

•  Normal operations monitoring  is a welcome
  addition in the context of Safety Management
  Systems
• The idea is simple, but the implementation is
  complex
• Must be useful to the airline safety management
• Must be adapted to the airline context
  (regulatory, social,)
• No unique solution exists
• The story behind the HFStG contribution

5
A much more complex story
The Safety Triangle
Extract from GAIN Asia-Pacific Regional
Conference, Tokyo November 14-15, 2002
6
The cast
MANAGEMENT
LINE PILOTS
7
The untold story
Flight Safety Officer / Safety Management System
MANAGEMENT
LINE PILOTS
OPERATIONS
Survey
Confidential reporting (ASAP)
Quality Assurance
Jump seat observations
Crew discussions
Contacts with instructor or CRM trainer
FDM (FOQA)
Incidents, accidents
Simulator or line checks debriefings
ASR
Training feedback
8
What is their message ?

• They speak about risk and PRECURSORS
• PRECURSOR ?
• Pilot practice
• Context
• System defense vulnerability
• Each is telling a different story from their
  perspective
• Frequency (FDM, survey)
• Details on operational context (informal
  feedback,  stories )
• On demand or always active

internal threats
Angel Performance vs. Natural Performance
9
A precursor The hole in the cheese
Error
Deviation
Latent condition
Element of context
10
A look at the cheese
Visibility

Sensors
on demand or always active
Accidents
Incidents
Investigation
D
LC
E
REX
C
P
FDM
P
E
ASR
LC
C
E
D
Survey
C
Operations / Production
Informal feedback Line checks, etc
LC
LC
LC
Requirements, Organization, Management
-
11
How does it work ? An example
 Aborted take-off due to flaps config error 
300 responses

• 79 similar cases detection thanks to C/L or
  parameters surveillance during taxiing

Context high workload during taxiign or
interruption of the task
Instantiation of the risk precursor ( E D
LC) ( C C ) ?Corrective action task to be
accomplished right after engine run-up
12
Some final ideasHow to make it work for you ?

• There is no such thing as A UNIQUE SAFETY method.
• You need to be aware of the strengths and
  weaknesses of your sensors (or lack of sensors).
• You need to address both management needs and
  pilot needs, because
• Management provides the resources to enable
  safety
• You cannot be safe without supporting your pilots
  to manage their own practices.
• Normal operations monitoring should be supported
  as a complementary objective, not as a specific
  tool (see next slide)
• because it supports the process to understand
  professional practices

13
Normal Operations Monitoring Suggested
principles

• 1. Structured method for collecting data on
  pilots activity during normal flights
• ? 2. Joint management / pilot association
  sponsorship
• ? 3. Voluntary crew participation
• ? 4. De-identified, confidential, and
  safety-minded data collection
• 5. Targeted data collection instrument
• 6. People accepted and trained, and
  calibrated for implementing the
  
  method
• ? 7. Trusted data collection site
• 8. Evaluation of data validity
• ? 9. Data derived targets for enhancement
• ?10. Feedback of results to the line pilots

14

• To share lessons learned
• To exchange on different national/cultural
  contexts
• To converge towards international cooperation

INVITATION
15
QUESTIONS ?