曝光臺 注意防騙 網曝天貓店富美金盛家居專營店坑蒙拐騙欺詐消費者

e) Human performance (pilot and controller capabilities, reliability, workload, errors,
etc.) must be included in any collision risk modeling.
f) Any assumptions of independence of systems, operations, or failure modes must be
thoroughly scrutinized. For example, could a third factor, such as inadequate
management, cause two, seemingly independent factors to be dependent or statistically
correlated?
g) Information from accident and incident reports, including operational error and pilot
deviation reports, should be used. These reports can yield qualitative, if not
SEPARATION SAFETY MODELING
4-2
quantitative, information on what can happen in the “tails” of “human performance
distributions.”
h) Consideration should be given to aircraft performance characteristics including wake
vortex generation and its potential effect on neighboring aircraft. Lack of
understanding of wake vortex phenomena may result in controllers and pilots applying
larger-than-necessary safety buffers in-trail to avoid negative wake vortex effects
during the approach phase of flight. Wake vortex encounters could well occur in
future en route airspace scenarios if separation minima (vertical or horizontal) are
reduced.
DISCUSSION:
The following paragraphs provide some of the background that led to these criteria. The
paragraph numbering corresponds to that above.
a) Safety modeling in en route airspace is quite complex, time consuming, resource
intensive, and data intensive. It is, therefore, wise to determine in advance what really
needs to be modeled. For example, if a rule is being considered that would allow
pilots to change course in a non-emergency without prior approval from Air Traffic
Control (ATC), it would be prudent to determine first whether such a rule would be
beneficial before embarking on an effort to include such a possibility in a safety model.
b) The decision to concentrate on en route airspace first is primarily due to the belief that
collision risk will be simpler to model in en route airspace than in terminal airspace.
Also, there are some ongoing efforts to model certain types of terminal airspace, such
as simultaneous parallel approaches, whereas we are unaware of any substantial effort
to model collision risk in en route airspace.
c) The beginnings of user-preferred routing are already in place with the FAA’s National
Route Program. It is expected that this philosophy of a more “free” airspace will be
expanded in the future. European airspace is expected to continue to have some fixed
routing, but it will also have some relaxation of fixed routing. Thus, it is necessary to
develop a model that can include various geometries of aircraft trajectories, as well as
changes in those geometries. It appears that this can be most easily accomplished with
a model that addresses aircraft-to-aircraft separation¾one that will be applicable
whatever trajectory geometries might exist.
d) It is the opinion of the team participants that a high level of airspace safety and
efficiency can only be maintained if some form of aircraft/airspace surveillance is
present, independent of pilots. It is also believed that, at present, this surveillance
cannot be adequately accomplished solely by means of on-board automated systems,
but will require oversight by air traffic controllers.
MODELING CRITERIA
4-3
e) Separation distances in en route and terminal airspace are much less than in oceanic
airspace. A controller error, pilot blunder, or malfunctioning avionics engenders a
much higher risk of collision when separations are small. Thus, in addition to
modeling the likelihood and effects of pilot and controller errors in initially causing a
loss of separation, the ability of pilots and controllers to recognize that a loss of
separation exists and to appropriately respond must be modeled, as must be the
responsiveness of aircraft in emergency maneuvers.
Human interactions with equipment must also be considered. For example, a blunder
might be the entering of incorrect information into a flight computer, or the response
to an automated warning might be inappropriate or delayed.
f) The high level of safety in the National Airspace System (NAS) is due to redundancies
in the system. For these redundancies to be truly effective they should be independent,
that is, the failure of one should not affect the probability of the failure of another.
Often redundant equipment, procedures, etc. are treated as being independent when, in
fact, the existence of another factor may impact all of them. Assumptions of
independence must, therefore, be scrutinized thoroughly when combining the
probabilities of system failures.
　
中國航空網 www.k6050.com
航空翻譯 www.aviation.cn
本文鏈接地址：a concept paper for separation safety modeling(15)