23RD ANNUAL CONFERENCE, Estoril, Portugal, 26-30 March 1984WP No. 59Review of Alpha Numeric Callsign Systems |
At the 1983 IFATCA Conference (Split) SC 1 was tasked with reviewing “Alpha Numeric” callsign systems. This resulted from a paper presented by SC 1 at Split, outlining the progress made in 1981/82 by the EARC Group, which was set up in 1979 with the following terms of reference : “To assist the ICAO Secretariat in the development of recommendations with the objective of the elimination of ambiguities in radio-telephony callsigns, including the assessment of an IFALPA proposal concerning an alpha-numeric callsign and its suitability for adoption in Annex 10”.
An alpha-numeric callsign is generally understood to be one in which the aircraft identification following the company designator is composed of both letters and numbers.
Many airlines use “flight- numbers” callsigns for commercial reasons. These consist of 1, 2, 3, or 4, digits following the company designator. In air traffic control communications, numbers are also used for headings, flight levels, SSR codes, pressure settings and RT frequencies, so that in any one RT transmission there can be a multitude of numbers. This leads to 2 problems. Firstly , callsign confusion. Secondly, number saturation. Both these situations are potentially dangerous and, in an attempt to reduce the amount of numerals contained in RT transmissions generally, different methods have been tried to substitute letters for numbers in callsigns, or to have a mixture of letters and numbers – the alpha-numeric systems.
Various alpha-numeric schemes have been introduced internally within MA’s airspace’s, and one particular scheme devised by IFALPA has been the subject of simulations in Canada and the Federal Republic of Germany. Operational trials of the IFALPA scheme have been carried out by 2 airlines.
Some countries have alpha-numeric systems operating at the moment, although in a purely domestic situation. In the UK, for instance, British Airways operates a ‘Shuttle’ (minimum check in, pay on board) service between London Heathrow and major UK cities. All flights in the ‘Shuttle’ operations have, for air traffic control purposes only , a ‘shuttle’ callsign which is used in RT transmissions and on flight data presentations, instead of using the company designators and flight number. This system was introduced in January 1981 in an attempt to reduce the number of similar ( and mainly British Airways) flight numbers on the same RT frequencies. In this respect, after some slight initial problem in the first two months , it has been operating successfully. Any attempt to revert to flight numbers for these flights would undoubtedly meet with stiff opposition from controllers. An example of the format used on flight progress strips and radar labels is ‘SH6M’ spoken as Shuttle Six Mike – this is instead of (perhaps) BA 6230 spoken, of course as Speedbird Six Two Three Zero. The ICAO company designator ‘SH’ is allocated to Servico Aero de Honduras, SA, who do not operate into the UK. One draw back is that it does involve an ATC assistant at Heathrow in converting the British Airways flight number callsign into its corresponding ‘Shuttle’ callsign on arrival at Heathrow, as the Airport Authority allocates aircraft parking stands based on the ‘commercial’ trip-number callsign as opposed to the ATC one. It is also necessary to ensure that the route charges are directed to BA and not to Serviceo Aero de Honduras, SA . This is taken care of by appropriate software.
France also operates a type of alpha-numeric callsign internally with their main domestic airline Air Inter. The detailed rationale of the operations is not known, but an example of the format is IT312BG. Once again, the flight number is replaced by an alpha-numeric form which, in the example above, would be abbreviated to ‘Air Inter Bravo golf’. It is understood that the system was introduced at the request of the ATS authorities rather than as a result of any airline requirement. This particular format will not be suitable for use after 1987 when ICAO 3 letter company designators will be used, since the total number in the flight identification would then exceed seven.
The IFALPA Scheme
In para 2.1.2 above mention was made of a scheme sponsored by IFALPA. This was proposed in 1972 and was based on the premise that since there are 26 letters in the English alphabet available for flight identification, as opposed to the 10 numerals (1-9 and zero) for trip numbers, there would be much less likelihood of similar callsigns occurring if letters were used instead of numbers.
The claimed advantage of this scheme over the previous mentioned ‘ad-hoc’ schemes is that it is based on mathematical principles, allowing direct conversions, either with tables or computer, between flight numbers and their alpha-numeric derivatives and vice-versa. It can also be used internationally , but it is not intended as a world-wide replacement for trip number callsigns. It is seen as a means available to reduce the possibility of callsign saturation in areas where this is a problem.
Two real- time simulations have been carried out using this type of alpha-numeric callsign. The first of these, lasting 2 days , was undertaken in Canada in January 1981, and the second took place in Germany during 6-12 )October 1982. Full reports were produced as a result of these simulations. Two operational trails have also been carried out. One, which commenced in February 1982 within the airspace of the UK , Ireland The Netherlands and Channel Islands, is still going on. It involves part of the fleet of UK Independent Airline, Dan Air. The other operational trial took place within South Africa, and involved South African Airways. This trial covered the period 1 April – 13 June 1983.
Simulations
The Canadian simulations took place at the Transport Canada Air Traffic Services and Experimentation Centre in Ottawa and was conducted under controlled laboratory conditions, whereby the proposed system of callsign identification was directly compared to a reference sample using present day formats. The ratio of aircraft flights using various approved and proposed callsign identification formats was changed from time to time at the discretion of the observers. However, to ensure direct comparison was possible throughout the exercise, the reference traffic sample always reflected the same ratio but under present day conditions. A typical terminal environment was selected for the experiment to ensure sufficient radio traffic and the air traffic controllers were members of the resident staff of the research facility.
The official report of the Canadian simulation concluded that :
- the results obtained from the a verbal debriefing of the controllers concerned, and analysis of audio tapes recorded during the exercise, indicated very few problems with the proposed callsign format or mixtures of various formats;
- the numeral in the proposed callsign format became a useful indication of separation needed in the approach/arrival situation;
- occasional errors in the addressing of aircraft with very similar callsigns occurred in both the present and proposed formats.
The German simulation was conducted in Munich using the simulator at the Air Navigation Services School there. The airspace selected was that under the jurisdiction of Frankfurt Approach Control. Traffic samples were modified for the task and prepared in 2 versions – the flight number callsigns and the corresponding alpha – numeric callsigns. Similar flight numbers were deliberately included for the assessment of possible ambiguities. A third version contained similar alpha-numeric callsigns. The trial was base on the assumption that all flight number callsigns would be replaced by their alpha-numeric equivalents, i.e. a total changeover to alpha- numerics (and aircraft registrations) with no flight numbers.
The 7 controllers participating in this simulation recommended the continued use of flight numbers. Nearly all foresaw problems when applying alpha- numeric callsigns and considered the risk of confusion to be greater than with flight numbers. The result of the simulation indicated that the flight number system is to be preferred to the new proposal, although the number of similar callsigns is reduced in the alpha-numeric system.
The German simulation reported that the decisive factors in reaching this conclusion were :
- more RT time was needed when using alpha-numeric callsigns;
- the new callsigns were more difficult to discern and remember, thus requiring higher concentration;
- alpha-numeric callsigns were more difficult to read on flight progress strips and radar screens;
- co-ordination was more elaborate and required increased concentration;
- the abbreviated form of alpha-numeric callsigns could not be applied. There were association problems between the abbreviated form and the full callsign, and similar short forms are more frequent;
- there was increased danger of confusion despite the reduced number of similar callsigns.
Operational Trials
The South African Airways (SAA) operational trial started on 1 April 1983. It involved primarily A300 and B737 aircraft crews. All international flights remained with the flight number system. Prior to the commencement of the trial extensive briefing operations were carried out. Almost 70% of the crews involved attended a briefing session, the remainder received briefing material. In addition briefings were given to 11 outstations involved. Air traffic control was represented on the briefing team. The trial was terminated on 13 June 1983. Separate reports were compiled by representatives of SAA and the South African Air Traffic Controllers association.
The airline report concluded that although the change was accepted quite well by most people concerned, resistance became more evident as the trial continued. Some specific points mentioned were :
- concern about similarity of some alpha-numeric callsigns with South African Air Force callsigns, which consist of one letter followed by 2 numbers and another letter. The military callsigns changed on an ad-hoc basis;
- where only the last 2 letters of the alpha-numeric form were used it was felt that it could be confused with state registered aircraft. (Note: It was stressed at the briefings that it was most important to use the SAA company prefix ‘Springbok’);
- it was felt that there would have been a far better chance of the alpha-numeric system being successful if the SAA operation included both domestic and international services. In this respect the airlines operations would be geared completely to alpha-numerics. More problems were encountered with B747 crews who rarely used the alpha-numeric system;
- The SAA Pilots Association ATS Committee is of the opinion that, given sufficient time, the alpha-numeric system, or a modified version of it could work. Since the trial was terminated a number of aircrew reported that they were just becoming accustomed to the alpha-numeric format when the trial ended. Many pilots were of the opinion that when the correct abbreviation was used the alpha-numeric system worked well.
The ATC report, based on the 10 week trial, indicated that all national ATC centres had been involved, and reports had been received from various FIR representatives. Its main findings were :
- the alpha-numeric system did not contribute to the elimination of ambiguity in RT callsigns to the degree it was intended;
- concentration levels on the ground and in the cockpit are increased;
- computer keyboard entry is more complicated and time consuming;
- similarities between various alpha-numeric callsigns also exist, still leading to confusion. This necessitates the continual use of the full form by ATC and the pilot, increasing RT time.
The other operational trial, which started in February 1982 is still running. It involves part of the fleet of Dan Air aircraft, an independent UK operator engaged in both scheduled and charter operations. The total Dan Air fleet comprises approximately 50 aircraft both turbo-prop and jet. The trial encompasses those Dan Air operations which fly within or between the UK, the Netherlands, Ireland and the Channel Islands airspace. Since the trial is still continuing no final report has been compiled. Various views have been expressed during the 21 months the trial has been running or, if none has been expressed, opinion has been sought. It is fair to say that there were difficulties in the initial stages. These fall broadly into 2 areas. In the operational area they were mainly the result of insufficient briefing prior to the start of the trial. In the administrative area, particularly with regard to route charges, the difficulties arose because the existing software programmes (designed to accept companies 2 letter ICAO designators followed by a number) would not accept what appeared to the computer to be a 3 letter designator. This resulted in manual processing of the alpha-numeric flights for invoicing purposes in the early stages. However, revisions to the software were eventually produced which solved this problem.
The extent of the trial has been severely limited by the fact that some adjacent foreign administrations have been unwilling to participate. This has meant that a fairly substantial part of Dan Air’s operation has been restricted to flight number callsigns. Among the comments received so far from more than one UK ATC Unit was that because the trial involved, in percentage off total movements at least , only a fairly small number of alpha-numeric callsigns (except at one or two airports) the trial had really not proved anything; there was only one way to give the trial a good test and that was by expanding it to encompass more than one airline and into more countries.
Opinion amongst air traffic controllers in the UK about the benefit or disadvantages of the IFALPA alpha-numeric system are varied. They range from those who consider the system has produced no benefits whatsoever ( or even made matters worse) to those who contend that the scheme has produced noticeable benefits. In fact the subject is ‘subjective’ in terms of whether controllers like or dislike it. This is also true as far as pilots are concerned, from comments received. The trial has produced no measurable benefits, but that is not to say there has been none. It has run for almost 2 years without major problems.
Specific areas of dislike concerned :
- the point that the order of letters and the number do not ‘roll off the tongue’ as figures tend to;
- the similarity on data displays (of various types) of some letters and numbers – notably I and 1, S and 5, Z and 2, B and 8 and O (Oscar) and 0 (zero).
ICAO Aspects
ICAO in its latest proposed revisions to Annex 10 – Aeronautical Telecommunications has amended the wording to paragraphs 5.2.1.6.2.1.1 Type (d) to allow for alpha-numeric callsigns as follows :
Present wording – the radiotelephony designator of the aircraft operating agency, followed by the flight identification number (no abbreviation allowed)
Wording effective from 7th June 1984 – the radiotelephony designator of the aircraft operating agency, followed by the flight identification (no abbreviation allowed)
Whilst undoubtedly made with good intentions, this revision could cause major problems if operators decide to adopt individual styles of ‘flight identification’. Unless guide-lines are established regarding what is an acceptable format for flight- identification there could be problems both in the operational field and in the area of data processing. Flight identification is, after all, the means by which ATC establishes and maintains contact with an aircraft to pass control instructions, and the amendment coming into effect in June could, if not ‘restricted’ by guidelines, allow free rein to operators to choose their own particular format of letters and numbers.
To conclude
On the evidence so far available it seems unlikely that it will be possible to develop a completely satisfactory alpha-numeric system that could be used as a direct and total replacement for the present flight number system world-wide. Certain alpha-numeric callsign systems have reduced number saturation and callsign confusion. In a limited operation these have worked well. Alpha- numeric systems should not be introduced by operators on a random basis. It should be carried out in an organised manner in agreement with the ATC Authorities concerned. In this context the proposed ICAO revision to the ‘type(d)’ format needs strict guidelines to prevent ad hoc alpha- numeric formats causing confusion.
The numeric character in the IFALPA format is not generally useful to ATC in the en route (Area) environment. It may be more useful in the approach/aerodrome phase. Any alpha-numeric format must be usable in its full form as well as any permitted abbreviated form (note: the ICAO amendment will not permit abbreviation of Type ‘d’ callsigns). Any method of abbreviations must be laid down in ICAO regulations, having been approved by the ATS authorities of member States. There could be problems with the display of data, regarding the similarity of some letters and numbers. In particular India/One, Sierra and Five, Zulu and Two, Bravo and eight, and Oscar and zero pairs that are potentially confusing.
It is recommended that:
That the following policy statement on alpha-numeric callsign systems be adopted :
“From a study of evidence available on the results of simulation and operational experience with the IFALPA and other alpha-numeric type callsign systems it is evident that none of these systems is suitable operationally for universal application as a total replacement for the ‘tripnumber’ type callsign system. The limited national use of alpha-numeric callsign systems may marginally reduce current RTF callsign ambiguity problems but, if authorised for use by international flights, ICAO Annex 10 must be specific as to the construction of these callsigns and such construction must permit their operational use both full and abbreviated forms”.
IFATCA recognises that the proposed revision to ICAO Annex 10, as described in para 2.5. of this WP is unsatisfactory and recommends that clear guidelines be established to prevent indiscriminate use of unsuitable ‘type (d) ‘ flight identification.
In association with IFALPA, recommendations for a new ‘type (f)’ callsign to cover the IFALPA alpha-numeric format, or a development of it, should be produced for possible inclusion in ICAO Annex 10.
Last Update: September 20, 2020