5
.
C
o
n
c
l
u
s
i
o
n
s
a
n
d
f
u
t
u
r
e
w
o
r
k
This report supports the flight movement inventory delivered to AERO2K partners. The work done for being able to deliver the final flight movement database was split into four phases. In the first phase, a prototype tool was developed in Access. Problems of slowness of the tool to run appeared quickly. The risk was that re-coding and modifying methods would demonstrate their inefficiency when increasing the amount of data. Consequently it was decided to migrate the prototype tool to an Oracle 9i database toolset. In the second phase, the methods previously developed for importing and merging data required some modification for fitting with Oracle and criteria used for identifying and cleaning flight data needed some reconsideration. For example, zone limitation criteria were re-evaluated. To facilitate the use and understanding of AERO2K flight movement tool in Access, a graphical user interface (GUI) was developed. This was not done in Oracle as the use of Toad software gave access to the database. In Phase 3, data collection appeared as a continuous and permanent process. Full 4-D trajectories were data not easily available, which required some work from the authority contacted. The difficulty was in finding the right person in the right organization. We hoped this would have been facilitated once the questionnaire would have been sent to regional ICAO offices. The acceptance of the questionnaire took so long that no data were collected this way. The questionnaire prepared in December 2001 was still not sent in March 2003. However, two of the goals set by the AERO2K partners i.e. the seasonality of air traffic and the obtaining of as many as possible real data trajectories were respected. Six weeks of data were collected and distributed through the year. In answer to ANCAT/EC2 (1992), new ATC based traffic movement data were included such as ETMS data for North America. The most significant omission from ATC-sourced data concerned China and FSU. Relationship established with Chinese and FSU authorities showed that long-term relation and face-to-face meetings were necessary; though this requires the allocation of a thinkable budget. Beyond these difficulties, having access to European and American data that cover the most important part of the world air traffic was a reasonable good achievement. The completion of the data with Back Aviation database enlarged the inventory to the world using a different concept that the attribution of a Great Circle line between city-pairs. The introduction of the concept of virtual day inventory based on trendline information from schedule data was also a novelty. Further statistical research for improving the method should be encouraged. In Phase 4, the goal of the project was reached and the uncertainties linked were expressed.
In summary, the objective of delivering a new civil air traffic movement database for 2002 based upon a new approach was reached. However further work linked to the intrinsic quality of the data would be required to gain in data quality and so emissions estimates. There is the need for an international effort to harmonize airports, aircraft and airline coding and for all parties to apply a unique code system. The increasing number of code-sharing flights leads to their duplication as the same flight appears under different callsigns. A recognized system for identifying code-sharing callsigns would need to be put in place by the airlines. The use of a unique identifier linked to the tail number for example would also facilitate greatly the identification of flights and the merge of trajectories. The automatic report of flight information would avoid data typing errors such as an “O” typed instead of a “0” and so would reduce the number of flights listed as inconsistent.
中国航空网 www.aero.cn
航空翻译 www.aviation.cn
本文链接地址：AERO2K Flight Movement Inventory Project Report(42)
