SUBJECT - MANAGEMENT OF FLIGHT SERVICE
SAFETY DEFICIENCY
Human factor issues in flight service centres are creating an
environment in which safety may be compromised.
FACTUAL INFORMATION
As a result of several occurrences, including the above occurrence
(199900266), the Australian Transport Safety Bureau, formerly the
Bureau of Air Safety Investigation, issued Safety Advisory Notice
(SAN) 19990055 to Airservices Australia on 29 July 1999. The SAN
was released to the public on 5 August 1999. The safety action in
the SAN suggested that "Airservices Australia should note the
safety deficiencies detailed in this document and take appropriate
action". While Airservices Australia advised the Bureau of receipt
of the SAN, it did not provide a formal response, nor was a
response required.
The Bureau has continued to monitor reported incidents for "failure
to pass traffic" and "failure to coordinate" that were considered
attributable to flight service. This monitoring revealed that for
the period 1 January 1999 to 1 December 1999 throughout Australia,
there were 21 occurrences of "failure to pass traffic" and 6
occurrences of "failure to coordinate".
Of concern is the increasing trend for "failure to pass traffic"
occurrences in the Sydney, Melbourne and Brisbane flight services
centres. In 1995 there were 4 occurrences, 1996 - 6 occurrences,
1997 - 6 occurrences, 1998 - 8 occurrences, and in 1999 - there
were 17 occurrences.
As reported in the SAN, human factors issues in flight service
centres are considered to be creating an environment where safety
may be compromised. These issues include, but are not limited to,
low morale, increased stress, increased workload, inadequate
supervision, inadequate rostering, fatigue and inadequate
resources. Airservices Australia's Safety Cell in Brisbane
highlighted these issues in an internal report prepared in July
1999. The report was forwarded to the Bureau on 28 October 1999.
The Bureau has also received advice that a Melbourne flight service
centre audit had similar findings.
Recent Occurrences
Summary of Occurrence 199904539 (ESIR 1999 02936 BCO)
The crew of VH-CCJ climbed to flight level (FL) 150 from FL140 to
avoid VH-TUZ, which was on a crossing track at FL140. Both aircraft
were operating under the instrument flight rules (IFR).
Consequently, the flight service (FS) officer was required to
provide traffic information to the crews, but did not do so.
The investigation found that immediately prior to the occurrence,
the FS5 air/ground operator became increasingly busy and, after CCJ
had reported at Kidston, that officer requested, and was provided,
the assistance of a coordinator. Multiple coordination tasks were
effected. CCJ was coordinated with FS1, however the transfer time
was incorrectly calculated and coordinated. The FS1 air/ground
operator did not notice the incorrect transfer time. The FS1 was
distracted by other events that were occurring and deferred
providing traffic information. This decision was made because the
FS1 believed that there was adequate time to re-assess the traffic
situation when CCJ was in his airspace. The crew of CCJ was alerted
to potential conflicting traffic on the aircraft's traffic alerting
and collision avoidance system (TCAS).
Local safety action following occurrence 199904539
Airservices Australia also investigated the incident. Their report,
dated 27 September 1999, recommended changes to the roster to
provide additional staff coverage, revised procedures for the
conduct of performance checks on flight service officers and
workstation modification to allow better visibility of the FS5
position from the operational control authority (OCA) position.
Workstation modification is expected in January 2000 following the
submission of a "Request for Change". The implementation of the
other recommendations was completed prior to 8 October 1999.
Summary of Occurrence 199904978 (ESIR 1999 03101 BCO)
The pilot-in-command of VH-KTJ sighted VH-SKZ pass 100 ft above his
aircraft. Both aircraft were IFR category and the flight service
officer was required to provide the crews with traffic information.
Neither crew was issued with traffic information regarding each
other's aircraft by flight service.
The investigation found that when SKZ taxied at York Island for
Cairns, the flight service officer assessed the traffic situation.
The officer relied on his memory of the proximity of the planned
tracks of SKZ and KTJ, rather than refer to his overhead chart. As
a result, the flight service officer made an incorrect assessment
and discounted the two aircraft as traffic.
The flight service officer reported that he may have been fatigued
due to insufficient sleep. Due to staff shortages the officer had
not had a 2-day break from the workplace in the previous 3-week
period. The officer may also have been distracted from his primary
task by the need for constant manipulation of the sensitivity
controls of the high frequency (HF) radio due to poor HF
conditions.
Local safety action following occurrence 199904978
The Airservices Australia investigation dated 11 October 1999
recommended that the group leader obtain ratings for the specialty
area to provide a degree of "insurance against staff being asked to
give up too many days off in their roster". The group leader was
rated in December 1999 and subsequently was able to provide
additional coverage on the roster.
ANALYSIS
While Airservices Australia acknowledged receipt of the SAN
19990055 dated 29 July 1999, there is no evidence that appropriate
action has been taken to address the concerns raised. Local
management has responded to individual incidents by conducting
investigations and has actioned some of the recommendations made
for local change. However, a strategic approach by Airservices
Australia is considered necessary to address the identified
systemic safety issues.
The increasing trend of "failure to pass traffic" occurrences
involves high-capacity fare-paying passenger transport operations.
The government regards the safety of fare-paying passengers as a
high priority. While a risk analysis may indicate that the chance
of a mid-air collision is relatively low, failure to address these
concerns within the flight service area of responsibility must
increase that level of risk.
The Australian Transport Safety Bureau (formerly BASI)
recommends that Airservices Australia address flight service
related issues that have the potential to seriously compromise
safety, including those relating to incidents where there was a
"failure to pass traffic" or a "failure to coordinate".
ATSB recommendation R19990220 was dispatched from the ATSB to
Airservices Australia on 19 January 2000. In accordance with sub
paragraph 12.15 of the ATSB(BASI)/Airservices MOU our response to
the recommendation is due on or before 20 March 2000.
HISTORY
Under the heading "Factual Information" recommendation 19990220
states that the ATSB issued Safety Advisory Notice (SAN) 19990055
to Airservices on 29 July 1999.
What is not mentioned as factual information is that this issue was
originally sent to Airservices as a formal recommendation some
months earlier than July 1999. At that time I spoke to (ATSB
officer) and complained that the data and the analysis of the data
did not support the conclusions or recommendation. (ATSB officer)
reviewed the recommendation and agreed the data did not support it.
The recommendation was then withdrawn by BASI and was never made
public in that form.
The BASI investigator responsible for developing the
recommendation, (ATSB officer), subsequently spoke to both (ASA
staff member), and myself about the original recommendation. I told
(ATSB officer) exactly why I rejected the recommendation and where
I felt the data and analysis were deficient.
In July 1999 BASI raised the issue in a slightly modified form as
SAN 19990055. As the ATSB has acknowledged, SANs do not require any
formal response. However, (ATSB officer) was again given a briefing
by (ASA staff member) about the actions Airservices were taking in
response to various flight service issues.
Without any further consultation with Airservices the ATSB once
again raised the profile of the issue to that of a formal
recommendation and sent it to Airservices on 19 January 2000.
Because of an intervening long weekend, the recommendation did not
arrive at Airservices until January 25 and was not seen by me until
the morning of January 27, the day BASI made the recommendation
public.
Had I seen the recommendation before it was made public by the
ATSB, I would have again, called into question the analysis of the
data and asked that the recommendation be withdrawn.
FACTUAL INFORMATION AND ANALYSIS
In the second paragraph of the section of the recommendation titled
"Factual Information" it is stated that in 1999 there were 6 flight
service "failure to coordinate" incidents. This is the only
reference to "failure to coordinate" incidents under the heading
"Factual Information". No other "failure to coordinate" figures are
given and no analysis is presented.
Airservices analysis of the "Failure to coordinate" data indicates
that the number of incidents Australia wide were 2 in 1995, 2 in
1996, 3 in 1997, 6 in 1998 and 6 in 1999. These sample sizes are
statistically far too small to allow any meaningful analysis, and a
figure of 6 incidents in 1999 conservatively represents less than
one one thousandth of one percent of all single IFR flights outside
controlled airspace for the year. These figures do not support the
argument that there is a trend or even a problem related to flight
service "failure to coordinate" incidents.
However, on the basis of a one year sample of six incidents, and
without any data analysis, the formal recommendation statement
requests Airservices, "address flight service issues that have the
potential to seriously compromise safety including those relating
to incidents where there was a "failure to coordinate".
Likewise, the second paragraph under "Factual Information", states
that Australia wide there were 21 "failure to pass traffic"
incidents in 1999. No where else in the document are any other
comparative figures given, or analysis provided, on an Australia
wide basis. Yet, partially on the basis of this data, and the data
referred to in the previous paragraphs, the "analysis" section of
the recommendation document states that Airservices should take a
strategic approach "to address the identified systemic safety
issues".
The data presented in the second paragraph does not support the
argument that there are any "identified systemic safety issues"
related to either "failure to coordinate" or "failure to pass
traffic" incidents.
In the third paragraph under the heading "Factual Information" it
is stated "Of concern is the increasing trend for "failure to pass
traffic" occurrences in the Sydney, Melbourne and Brisbane flight
service centres." The figures are presented collectively and state
that in 1995 there were 4 occurrences, 1996, 6, in 1997, 6, 1998, 8
and in 1999 there were 17 "failure to pass traffic" incidents.
Strangely, having identified Melbourne as contributing to the
"increasing trend" in "failure to pass traffic" incidents, the list
of incidents provided by the ATSB does not mention Melbourne.
Even though the ATSB and Airservices total number of incidents
agree, the distribution of incidents does not. The reason for this
is that, the ATSB have listed the incidents by where they occurred,
rather than by, which flight services centre was involved.
Airservices have listed with respect to which centre was providing
the service at the time of the incident.
Collectively, the numbers of "failure to pass traffic" incidents
recorded by Airservices for Sydney, Melbourne and Brisbane in 1995
is 7, 1996, 10, 1997, 9, 1998, 3 and 1999, 12.
While the collective spike in 1999 is well worth closer analysis,
neither the ATSB or Airservices figures allow any meaningful
conclusion that there is any "increasing trend for "failure to pass
traffic" occurrences". This point becomes even more evident when
the figures are separated and presented independently for each of
the three centres.
Separately the "failure to pass traffic" occurrence figures
are
(Airservices data)
1999 1998 1997 1996 1995
SYD 6 0 3 4 4
MEL 1 1 0 3 0
BRN 5 2 6 3 3
TOT 12 3 9 10 7
To those without any formal knowledge or qualifications in
statistical analysis, an increase from 4 to 6 may look like an
increasing trend. However, in reality, none of the figures provide
any meaningful evidence to support the argument that there is an
"increasing trend" in "failure to pass traffic" occurrences
attributable to flight service in operations. Even if the ATSB
figures were used, the pattern would not change.
In the fourth paragraph under the heading "Factual Information",
the recommendation report refers to human factors issues, and two
Airservices internal reports that refer to issues such as, low
morale, increased stress, increased workload, inadequate
supervision, inadequate rostering, fatigue and inadequate
resources.
The entire thrust of both the recommendation and the supporting
information, is that, there is a direct relationship between the
number of flight service "failure to coordinate" incidents and the
"increasing trend" of failure to pass traffic incidents, and the
reported poor morale, etc, within flight service. However, as
already pointed out, there is no evidence of increasing numbers of
"failure to coordinate" or "failure to pass traffic incidents".
Further, an analysis of the 12 "failure to pass traffic" incidents
in 1999 reveals that only 2 of the incidents are attributable to
the types of issues mentioned in the Airservices internal reports.
One of the incidents was directly attributed to high workload,
while the other was attributed to high workload, fatigue due staff
shortages and despondency due impending redundancy.
The data presented does not support the argument that there is any
specific causal relationship between the group of incidents
recorded in 1999 and human factors issues, such as low morale,
within flight services.
HUMAN FACTORS ISSUES AND FLIGHT SERVICE OPERATIONS
As part of the reform of Australian airspace, a decision was made
in the late 1980s to eliminate flight service as a separate
function with some of the duties ultimately being transferred to
air traffic control. This decision clearly has employment
implications for some flight service officers.
The general morale of many flight services officers is
understandably low as many of them have been facing the possibility
of employment termination and prolonged uncertainty about when the
terminations may be enforced. As earlier demonstrated, there is no
major problem, or trend with the number of flight service
incidents, and while there is the potential, there is in fact no
direct causal relationship evident between reported poor morale
within flight service and the incidents sighted in support of the
ATSB recommendation.
ATSB recommendation 19990220 "recommends that Airservices Australia
address flight services related issues that have the potential to
seriously compromise safety". The recommendation implies that there
are in fact safety issues to be dealt with and that Airservices is
not addressing them.
Airservices completely rejects the ATSB assertion and is satisfied
that the flight service issues are human resources industrial
issues rather than safety issues and are being dealt with
appropriately by local and national management.
By 23 March 2000, the existing, separate flight service function
will be withdrawn from the Canberra/Ballina corridor thereby
effectively removing flight service from the entire east coast of
Australia. By mid June 2000, flight service as we know it will
cease to exist. A small number of flight service officers (20 to
30) will be retained in Perth until mid 2001 while Airservices
works towards completing the integration of directed traffic
information (DTI) into the new TAAATS environment.
SUMMARY OF RESPONSE
In accordance with sub paragraph 12.17 of the
ATSB(BASI)/Airservices MOU, Airservices Australia formally rejects
recommendation 19990220 on the basis that the analysis of the data
presented is flawed, and the recommendation directly implies that
major safety issues exist within flight service, and that
Airservices management is not taking appropriate action to resolve
those issues.
The data presented in support of the recommendation does not
support the arguments that there are;
- a high number of "failure to coordinate" incidents
- an increasing trend of "failure to pass traffic" incidents
- any direct causal relationship between some existing flight
service morale issues and the data presented
- identified system safety issues.
Airservices is satisfied that the actions it is taking in response
to the various human resource issues currently existing within
flight service is appropriate and while safety issues have been,
and will continue to be monitored, the safety of operations has not
been compromised.
The following correspondence was forwarded to Airservices
Australia on 17 March 2000.
I refer to your letter to (ATSB officer) of 28 February 2000
concerning Recommendation R199900220.
In accordance with established practices, each of the points made
in your letter has been examined against the information used to
develop the recommendation. We accept that there are differences in
interpretation of the "facts" and that the statistical data is not
necessarily compelling of itself. However, the potential
consequences of the incidents are significant. It is clear that the
intent of the recommendation, to raise the profile of the safety
issues in flight services centres, has now been achieved.
As recently discussed by (ATSB officer) with you in relation to
another failure to provide traffic information, the actions taken
by Airservices are reflected in the description of local safety
action in the Occurrence Brief:
"Airservices Australia advised the ATSB that the existing flight
service operations will continue until early 2001 when the nation
program of incorporating directed traffic information into TAAATS
will be completed. Airservices is aware of the need to be vigilant
in the management of this change process."
A copy of the Occurrence Brief 199903790 is attached for your
consideration prior to its release for interest party
comment.
In view of the developments, you may wish to advise whether
Airservices' rejection of the recommendation stands.
I refer to your letter of 17 March 2000 concerning ATSB
recommendation R19990220.
I would be interested to know which "facts" you believe the ATSB
and Airservices have interpreted differently? The only "facts"
presented were the number of incidents, and on this matter we are
in agreement.
Your statement that the "statistical data is not compelling of
itself" is a gross understatement. The comparative number and
spread of incidents sighted provides no statistical basis
whatsoever for the conclusions drawn by the ATSB. Pressing home
your argument by stating "However, the potential consequences of
the incidents are significant" simply indicates a lack of
understanding of contemporary risk management practice.
The potential worst case consequences for some of these incidents
are significant. However, the events themselves prove that the most
likely consequences of most of the incidents are not significant.
Also, the relative frequency of these incidents is so low as to be
almost immeasurable. Contemporary risk management is not only based
on consequence, but also relative frequency. In most cases, for a
risk that does not have an immediate outcome (i.e an incident
rather than an accident) to be considered intolerable, it must not
only have significant consequences, but also moderate or high
relative frequency.
The number of flight service incidents has not changed over the
last 5 years and is still so low that is practically
immeasurable.
If the ATSBs position is that likely consequence and relative
frequency are not generally important then almost all incidents
should elicit an ATSB safety recommendation as the potential worse
case scenario for most incidents are significant.
The last sentence of the second paragraph of your letter is
self-serving and incorrect. The awareness of potential safety
issues within flight service was always high. R19990220 has not
raised the profile of the safety issues in flight services centres
at all. Unfortunately, what it has done is seriously erode the
ATSBs professional credibility within Airservices. Airservices is
still of the view that the flight service case presented by the
ATSB is flawed and does not support the conclusions or the
recommendation.
In response to the local safety action statement included in
Occurrence Brief 199903790, I did not have the opportunity to read
it before it was released to the public. In essence, the statement
is correct, but could be misleading. Stating that "existing flight
service operations will continue until early 2001" could be
interpreted as meaning that there will be no change to flight
service operations until early 2001.
As you would now be aware, except for a small pocket of air space
around Canberra and Sydney, flight service has been removed from
the entire east coast of Australia. The Sydney Canberra flight
service function will be removed by the end of April and the
majority of the remaining non "J curve" flight service function
will be discontinued by June. The last traditional flight service
function will operate in Western Australia with a total of 20 to 30
officers until approximately January 2001.
With respect to the last sentence in your letter, and based on the
arguments contained in my letter to (ATSB officer) of 28 February,
2000, I confirm that Airservices formal rejection of recommendation
19990220 stands.
