Recommendation

SUBJECT - ASYMMETRIC TRAINING AND CHECKING





INTRODUCTION - REGIONAL AIRLINES SAFETY STUDY



Between October 1995 and July 1997, the Bureau of Air Safety
Investigation undertook a study of the safety of Australian
regional airlines. The objectives of this study were to:



(a) identify safety deficiencies affecting regional airline
(operations in Australia; and

(b) identify means of reducing the impact on safety of these
deficiencies.



For the purpose of the survey, regional airlines were grouped
according to the number of passenger seats fitted to the largest
aircraft operated by that airline in January 1997. The groups are
defined as follows:



(a) Group 1: 1-9 seats;

(b) Group 2: 10-19 seats: and

(c) Group 3: more than 20 seats.



The study involved analysing data obtained from:



(a) responses to a survey of Australian regional airline
employees;

(b) discussions with Australian regional airline employees and
managers;

(c) air safety occurrence reports involving regional airlines over
a 10-year period (1986-1995) from the BASI database.



This Safety Advisory Notice addresses one of the safety
deficiencies identified as a result of this study.





SAFETY DEFICIENCY



Some asymmetric training is being conducted in a manner or under
conditions where the risk involved is not commensurate with the
training value gained.





FACTUAL INFORMATION



Survey results



Pilots were asked to outline any flying training exercises that
caused them concern. Sixty-nine pilots (18%) responded to this
question. The most common responses were as follows:



Forty-seven pilots (68% of those answering this question) were
concerned about the dangers of asymmetric training in general. The
most common concern was that simulated asymmetric flight (flight
with one engine set at or near a zero-power setting) was being
conducted close to the ground, or in aircraft configurations or
flight conditions that increased the risk of an accident.



Twenty-three of the respondents who commented on the danger of
asymmetric training, said that "V1 cuts" (simulated failure of an
engine at V1 speed) were excessively dangerous due to the marginal
performance of the aircraft on one engine, or due to the manner in
which this exercise was practised. Five respondents cited examples
of V1 cuts being practiced at night and one described a V1 cut
while the pilot was wearing "foggles" (devices to restrict the
pilot's view to the instrument panel). Many pilots felt that V1
cuts should only be practised in flight simulators.



Examples of survey responses



"During a twin engine, full flap go-around, [the training captain
simulated] an engine failure when power [was coming] up, leaving
the aircraft hanging on one engine low to the ground. With or
without the [landing] gear, the aircraft will not climb [on one
engine] with full flap selected."

- Pilot, respondent 465



"[I am concerned about] Vmc demonstrations and practice. [They
are] all very dubious demonstrations at this level of operation and
definitely a safety hazard in practice for no gain in
safety."

- Pilot, respondent 472



"[I am concerned about simulated] engine failures at Vr (rotate
speed) at night. Besides being a breech of the operations manual
and CAOs, there is no margin for error in such an exercise."

-Pilot, respondent 493





Asymmetric training and checking requirements



Asymmetric training is a significant aspect of the multi-engine
training syllabus and flight test. Pilots are also required to
undergo training and demonstrate proficiency in asymmetric flight
to obtain and renew multi-engine instrument ratings in addition to
other required periodical flight checks.





Aircraft performance



Many of the aircraft types used by regional airline operators
demonstrate limited performance during asymmetric operations and do
not have to meet the "one engine inoperative" performance
requirements that must be met for certification of larger
transport-category aircraft. This effectively means that there is
an accepted period of risk during the take-off and the initial
climb phases of flight. If an engine failure were to occur during
this phase of flight, climb performance would not be guaranteed.
Notwithstanding, and as part of training and checking activities,
pilots are expected to demonstrate proficiency in asymmetric
operations during all phases of flight, including takeoff and
initial climb.





Related occurrence and safety action



An accident, in which two pilots were killed in a Fairchild Metro
II during a night training exercise at Tamworth, highlighted the
hazards associated with the conduct of some asymmetric training
exercises, particularly V1 cuts below 1,500 ft AGL at night. The
Bureau issued two interim recommendations as a result of that
accident which highlighted similar safety concerns that have
emerged from the Regional Airlines Safety Study.



Interim Recommendation 950224, issued on 29 April 1996, stated
that CASA should amend the Civil Aviation Regulations and Civil
Aviation Orders to ensure that "when a provision of the
Aeronautical Information Publication specifically prohibits certain
manoeuvres and procedures, then the prohibition has legal force
which is reflected in relevant Civil Aviation Regulations and Civil
Aviation Orders". BASI also recommended that CASA take "appropriate
steps to inform and educate the industry on the hazards involved in
asymmetric training operations in conditions of low visibility and
at night". CASA indicated that the Regulatory Structure Validation
Project would rectify the problems identified in the BASI
recommendation and that an article would be produced and included
in the CASA Flight Safety Australia magazine. To date, BASI has not
been notified of any action taken in response to this
recommendation.



Interim Recommendation 960035, issued on 29 July 1996, stated that
CASA should "address the issue of endorsement training requirements
for aircraft above 5,700 kg MTOW where a simulator is not
available". BASI also recommended that the review of the
multi-engine training syllabus should cover the possible difference
in flight training when a simulator is used for training and for
the conduct of emergency procedures. CASA has since issued Civil
Aviation Advisory Publication 53.1-(0) which provides detailed
guidance on the conduct of multi-engine endorsement training and
reiterates that multi-engine training at night should be conducted
in accordance with Aeronautical Information Publication OPS-77. In
its response to the interim recommendation, CASA stated that while
it had no legislative power to mandate the use of simulators, the
new syllabus encouraged the use of an approved type simulator for
the conduct of endorsement training. CASA added that the need for
different training requirements, depending on the availability of a
simulator, was currently under review. However, CASA indicated that
it would prefer to have a generic syllabus for multi-engine
training.





ANALYSIS



Most asymmetric training conducted by regional airline operators
must be undertaken in aircraft as flight simulators are currently
available for only a few of the regional airline aircraft types. In
addition, many regional airline aircraft demonstrate limited
asymmetric performance, even at training weights. Therefore,
asymmetric training can be a potentially hazardous activity if it
is not conducted in accordance with accepted safe practices that
are reflected in Australian and international regulations and
guidance material.



While there is some benefit in presenting a realistic scenario
when conducting asymmetric training, the risks must be weighed up
against the training benefits. There may also be insufficient
allowance being made for the difference between flight manual
performance expectations and the aircraft's actual performance.
Consideration should be given to conducting the same training at a
height above ground level that would provide a greater margin for
error in those cases when aircraft may be operated inappropriately
or when aircraft performance does not meet expectations.



The reports provided in the survey suggest that some training and
checking pilots may be presenting unrealistic asymmetric scenarios
during training and checking exercises. Such an approach further
diminishes the safety margins available as well as the capacity of
training and check pilots to recognise and recover from developing
unsafe situations.