Summary

With the advancement of technology in Air Traffic Control, some Air Navigation Service Providers (ANSPs) have removed the approach provision from their co-located manned towers and have moved them into Area Control Centres (ACC). Commercial and economic pressures continue to drive ANSPs to maximise the number of aircraft controlled with the minimum number of staff. This consolidation could require Air Traffic Controllers (ATCOs) to exercise multiple ratings during their operational shift, for example approach and area.

This paper considers the implications of exercising more than one rating during an ATCOs operational shift, including fatigue, safety risks, and the impact on human performance.

Introduction

1.1. Some ANSPs have removed the approach provision from their respective towers and relocated them into ACCs. With more surveillance-rated ATCOs located at a facility, ATCOs could be retrained on the ratings that they do not hold, and then be required to move between providing an approach surveillance service (APS) and then an area control service (ACS), throughout their operational shift, thereby utilising their two different ratings.

1.2. This kind of combination is understandable and somewhat expected in the environment of reducing costs borne by the ANSPs. The airlines and regulators will continue to demand lower costs and the efficiencies that would drive that.

1.3. However, often there is a tension between efficiency and risk. This paper will examine some of the potential risks, including both operational and fatigue risks associated with ATCOs exercising multiple ratings during one operational shift.

Discussion

Reasons for co-locating approach and area sectors

2.1. ANSPs are expected to remain competitive, as any other business would be required to, and explore methods of reducing costs and therefore remaining efficient and competitive.

2.2. Co-locating multiple approaches and airspace sectors into a single facility may allow for favourable cost-saving benefits for the ANSP.

2.3. Relocating approaches into facilities could allow for the introduction of new technology for airports that did not have surveillance approaches provided at the towers. This could be considered as an improvement in safety and efficiency for the airport concerned.

2.4. Some of the airports that have their approaches combined into one facility could have been difficult to staff due to their distances from large metropolitan areas, where generally, facilities are usually located.

2.5. Any update or improvement of the technology in use can be quickly implemented for numerous approach sectors if they are co-located within a single facility.

2.6. Some ATCOs might see an increase in job satisfaction and the potential to alleviate boredom during periods of low traffic. The potential to move between sectors that require different ratings might appeal to some controllers especially when their sectors, or airports, can become quiet at times.

2.7. Another example of possible cost savings, and one that this paper is more interested in, is the potential to utilise the staff more effectively during periods of persistent or unexpected staff shortness. ATCOs could be required to retrain for the rating they don’t hold and, once completed, could effectively utilise both ratings during the same operational shift could allow for increased resilience against unexpected staff shortages, or even allow for a reduction in the number of ATCOs that the ANSP believe will be required to maintain the operation.

ATCO ratings

2.8. ICAO defines a rating as:

ICAO Annex 1 – Definitions

2.9. The Air Traffic Controller ratings, as described by ICAO Annex 1 are (EASA Aerodrome ratings are ADV (Aerodrome Control Visual) and ADI (Aerodrome Control Instrument)):

ICAO Annex 1 – 4.5.1 ATCO Ratings

2.10. To obtain a rating, an ATCO must demonstrate a level of knowledge, experience and skill appropriate to the privileges granted. Each rating differentiates between the skill, knowledge and experience required to be able to effectively provide the level of service expected by aircraft operating within the airspace, or within the vicinity of an aerodrome (ICAO Doc 10056 Air Traffic Controller Competency-based Training and Assessment provides example rating training syllabi for all ratings).

Maintaining competency

2.11. ICAO Annex 1 states that:

ICAO Annex 1 – 4.5.3.4 Validity of ratings

2.12. Within Europe there is a requirement to have a unit competency scheme where it must outline the maximum continuous period when the privileges of the unit endorsement are not exercised during its validity. This period shall not exceed 90 calendar days (Regulation (EU) 2015/340 ATCO.B.025 (a) (2)).

2.13. It also must state the minimum number of hours for exercising the privileges of the unit endorsement within a defined period of time, which shall not exceed 12 months (Regulation (EU) 2015/340 ATCO.B.025 (a) (3)).

2.14. These figures are the least restrictive to maintain the competency of the ATCO and ANSPs can adjust the minimums and maximums to ensure all ATCOs are able to maintain their competency. One example of this would be the requirement for an ATCO to have 14 hours exercising the privileges of their licence within each calendar month.

2.15. The holder of multiple ratings must be able to meet the requirements to maintain their competency on all the positions that they operate. Administrative and operational demands on the ATCO, such as simulator training, OJT and the necessity to take annual leave, make the ability to meet the hours as required by the competency schemes, quite challenging. Even if the ATCO does manage to meet the minimum hours as required, ATCO might not feel confident that they have had enough exposure to all the nuances of their sector(s) and as such find that they are not maintaining the radar skills they would like.

Operational considerations

2.16. There are no restrictions on an ATCO using different ratings during their operational shift and it is not uncommon for ATCOs to exercise different combinations of ratings. For example, where an approach function is still located at a tower, an ATCO is often required to move between aerodrome control and approach during their operational shift.

2.17. So, there are scenarios where ATCOs can safely exercise multiple ratings in one operational shift. However, each situation has to be considered on its merits. The following is a list of considerations for exercising multiple ratings in a surveillance environment.

Surveillance ranges

2.18. ATCOs moving between ratings with differing surveillance ranges could misinterpret the horizontal separation that they are providing between two aircraft.

Separation requirements

2.19. Different control positions may have different minimum surveillance separation minima. This could result in the risk of an ATCO applying the wrong separation minima between aircraft.

Aircraft and pilot behaviour

2.20. Approach and area controllers have different knowledge and skillset requirements. Approach controllers will often be working with aircraft flying slower, lower and in denser air, with the ability to have a smaller radius of turn. Pilots will usually be operating with a sterile cockpit (What Is a Sterile Cockpit and Why Does It Matter? https://simpleflying.com/sterile-cockpit/) below 10,000ft and their response times to radiotelephony instructions are likely to be rapid.

2.21. Area controllers tend to work aircraft at higher levels climbing, descending to, or already at their cruising level. Aircraft will therefore have a much faster ground speed, operating at or near their operational limits, with much larger rates of turn. There is less need for a sterile cockpit and as such radiotelephony transmissions are more prone to be missed or delayed.

2.22. These differences could result in an ATCO expecting something from the pilot or the aircraft which doesn’t materialise. Aircraft manoeuvre differently depending on factors such as location, altitude and speed and this is evident on surveillance screens. Reaction and resolution times are different and ATCOs that transition from using one rating to another must be aware of the potential human error when assessing things like aircraft closure rates, turn radii and radiotelephony response times.

Technology

2.23. Surveillance displays, air-ground and ground-ground communication systems and information systems can often differ between approach and area. This may result in some confusion or stress with the ATCO being unable to use the new technology or systems as effectively as only being trained on one rating.

Geographical location

2.24. The geographical location of the area and approach sectors are likely to be different. This could lead the ATCO to relate the conditions (such as the wind strength) of the previous sector to their new sector.

Rules and procedures

2.25. Differences in rules or procedures between sectors that require the use of different ratings have the potential to cause the ATCO to misapply them when moving from one sector to another. Something like frequencies, that could be similar, could result in the controller incorrectly using a frequency associated with the previous sector.

2.26. A number of these considerations also apply to controllers who operate on only one rating but with two or more sectors, however, arguably to a lesser degree. For example, an ATCO rated on two different approach or area sectors. They are likely to have different rules and maybe different systems that the ATCO uses. However, there are also numerous similarities such as the possibility of using the same, or similar surveillance ranges, aircraft having similar rate of turn and resolution and reaction times are likely to all be similar. Each situation would have to be considered on its own merits taking into account these considerations.

Concurrent operation of two ratings

2.27. Another scenario with ATCOs who hold approach and area ratings is the possibility to merge an approach sector into an area sector. An ATCO could combine an APS position with the ACS position they are currently operating.

2.28. There would be a need to have similar technology and numerous ANSPs are utilising this method of operation already. Operating ACS and APS concurrently could possibly exacerbate the likelihood of the operational hazards discussed in this paper with the ATCO having to shift
their cognitive mental model from one rating to another during live operations.

2.29. The ability to combine two ratings will allow for the reduction of one controller which could allow for a reduction in safety nets. This was discussed appropriately within the review of the Four Eyes Principle (4EP) paper in Bali, 2013.

2.30. The paper on 4EP introduces the concept of the Two Eyes Principle (2EP) which is a scenario which could be envisaged here when allowing the combining of approach with area. This method might be utilised to allow for a persistent or temporary staff shortage or when the traffic demand allows for it like night shifts, for example.

2.31. If a controller is carrying out the roles of two people, then there is an increased risk that the attention needed by carrying out one task may lead to a failure of detecting an error in the other task (IFATCA Working paper – Review of Single Person Operations and Four Eyes Principle, Bali, 2013).

2.32. However, if the combination of ACS and APS is managed appropriately, and the procedures are designed carefully with the controller in mind, workload monitored and controlled, and training is sufficient, then a contiguous and coherent volume of airspace could be created, allowing the controller to develop a single mental model of the situation. This scenario is described in the working paper on the Study of the Remote Towers Concept, Gran Canaria, 2014 (IFATCA Working paper – The Study of the Remote Towers Concept, Gran Canaria, 2014 2.10.5).

2.33. This combining of sectors is normally only seen within the surveillance environment, however, there are also towers that operate APS concurrently with ADI. Under specific conditions ATCOs in the tower are able to provide a combined APS and ADI service. Through advances in technology, RiTT (Radar in the tower) has been operational at Belfast, Cardiff and Southampton in the UK for many years (Retrieved from https://www.nats.aero/news/radar-tower-operations-trials/).

2.34. IFATCA has policy (IFATCA Technical and Professional manual, WC10.3.2) on this procedure which clearly states that:

IFATCA TPM – WC 10.3.2 Work and Rest scheme

2.35. The combination of approach and aerodrome is considered unacceptable as it encourages single or lone-person operations and should be strongly discouraged.

2.36. Within a surveillance operation, it is unlikely that an ATCO will be operating by themselves, however, if the ATCO becomes distracted or the workload increases unexpectedly, they could find that their ability to manage the combined ratings is rather more difficult.

Mode Switching

2.37. Mode switching occurs when you use more than one mental model to perform the same task. This can be seen, for example, when you drive a car in another country where you drive on the opposite side of the road, or when you use an unfamiliar computer with a different keyboard layout. In these examples, the task is the same, however, it requires more thought and caution when switching between them.

2.38. Mode switching is a familiar occurrence for pilots who hold multiple type ratings (and has been a causal factor in some aircraft incidents (Air Nippon report https://aviation-safety.net/wikibase/138837)). And it is becoming more apparent within ATC due to the emergence of new technologies, such as electronic flight progress boards.

2.39. Two Human Factors specialists, Zsófi Berkes and Miguel Aulet, who have previously worked at NATS in the United Kingdom, determined two types of mode switching and described them in EUROCONTROL’s Hindsight magazine (Hindsight 28th Edition https://www.skybrary.aero/bookshelf/books/4508.pdf).

Change-related mode-switching: this takes place as new procedures, tools or airspace are being developed. Controllers operate a new tool (e.g., electronic strips) or new airspace in the simulator, and then afterwards have to operate a live traffic position operating with current tools and procedures.

In-service mode-switching: occurs when controllers switch between systems in live operations or when controllers switch between sectors or roles.

2.40. An example of a mode-switching error in flight is the Air Nippon 737 incident, in Japan back in 2011. The First Officer was attempting to open the flight deck door with the door release switch located on the central pedestal. The First Officer instinctively went to the switch position that was located in a different 737 variant and inadvertently applied trim to the rudder, causing the aircraft to lose altitude.

2.41. Mode-switching errors in the ATC environment are more often associated with an annoyance or a distraction, but it has the potential to increase the likelihood of an operational error, like the application of incorrect separation standards. People can, and do, regularly make small mistakes when mode switching and might not either be aware of, or overly concerned, due to their capacity to manage such errors and as such, these can often go unreported.

2.42. When switching between tasks, if the tasks are similar, like switching from driving on the left of the road to driving on the right, it is much easier to succumb to mode-switching errors. However, if the task is vastly different, such as driving a car to the airport and then flying an aircraft, then the tasks are so dissimilar that there is less risk of accidentally adopting an action from a previous task.

2.43. There are some examples where mode switching is actively managed. Air Services Australia will not allow a controller to return to live operations on the same day after being involved in any simulated activity. Another ANSP within Europe will only allow for an ATCO to be moved once, from one rating to another during their operational shift.

2.44. Mode switching is only one of many factors which impact the cognitive capacity of the ATCO. At conference in Las Vegas, 2016, the working paper on the Cognitive Processes in Air Traffic Control describes the various influencing factors. These can include distractions, complex systems/procedures, increasing workload and ageing.

Ageing

2.45. At conference in Punta Cana, 2010, the working paper on The Ageing Controller describes the process of cognitive decline of an ATCO in more detail, it points out that the 5 cogitative skills that are likely to decline as we age are:

• Multitasking,
• Direct attention to information source,
• Managing working memory,
• Active problem-solving and
• Diagnosing novel situations/problems.

2.46. Whilst there are multiple cognitive skills which are seen to actually improve with ageing, such as the ability to develop plans, it was felt required to introduce a policy to allow for an ageing ATCO to reduce the number of ratings/endorsements they hold to a reasonable minimum.

2.47. An excerpt from current IFATCA policy (IFATCA Technical and Professional manual, WC10.3.7) is:

IFATCA TPM – WC 10.3.7 Ageing ATCOs

Fatigue

2.48. Fatigue is acknowledged as a hazard that degrades various types of human performance and can contribute to incidents and accidents. Fatigue is inevitable in a 24/7 industry. Fatigue cannot be eliminated; therefore, it must be managed (ICAO Doc 9966 – Manual for the Oversight of Fatigue Management Approaches).

2.49. IFATCA defines fatigue as:

IFATCA TPM – Definitions

2.50. A common perception equates fatigue with a feeling of being sleepy or tired. However, tiredness may refer to the ability to initiate sleep while fatigue refers to the inability to maintain job-sufficient alertness (IFATCA Working paper – Fatigue Management in Air Traffic Control, 2.2.3, Gran Canaria, 2014).

2.51. ICAO Annex 11 (ICAO Doc 9966 – Chapter 2, section 2.28 Appendix 6 and Appendix 7) – Air Traffic Services, contains the Standards and Recommended Practices (SARPs) which are related to fatigue management. It permits States to choose whether to develop regulations that prescribe scheduling limits or to authorise air traffic service providers to use a fatigue risk management system (FRMS) to manage fatigue.

2.52. Fatigue Risk Management Systems (FRMS) are designed to provide a data-driven means of continuously monitoring and managing fatigue-related safety risks. Systems must be predictive, proactive and reactive. In essence monitoring current and future working practices and considering the implication on fatigue (ICAO Annex 11 – Appendix 6, FRMS requirements).

2.53. These two distinct methods (ICAO 9966 Chapter 1, section 1.1 Fatigue management approaches supported by ICAO SARPS) for managing fatigue share two important factors. Firstly, they are based on scientific principles, knowledge and operational experience. Secondly, fatigue management has to be a shared responsibility between the State, service providers and individuals.

2.54. Utilising either method, service providers will be required to prescribe maxima for work periods, minima for non-work periods and other elements as relevant to the risks associated with a specific type of work.

2.55. These basic principles relate to the need for sleep, sleep loss and recovery, circadian effects on sleep and performance, and the influence of workload.

Workload

2.56. Workload can contribute to an individual’s level of fatigue. ICAOs definition of fatigue describes workload as a “mental or physical activity” and recognises workload as a potential cause of fatigue.

2.57. It is generally accepted that individuals cope best with moderate levels of workload. ATCOs in particular work best when workload is medium to medium-high and human error tends to occur at the extreme ends of the workload spectrum. Low workload tends to expose issues with fatigue where periods of high workload can introduce risk (Retrieved from https://think.aero/views/challenging-the-80-20-fallacy-of-atco-workload/).

Relationship between workload and performance – De Waard 1996

2.58. As the task demand increases, performance improves, while the effort to provide the performance goes down (sections D and A1). Then there is an optimal performance area (A2) where it is levelled to an optimum, even if the task demand increases while the effort is constant. In section A3, as the demand continues to grow, the human effort to keep the performance goes up as well, until a certain limit (section B) where the performance drops.

2.59. A controller’s susceptibility to high workload can vary depending on multiple factors and is never a constant. A controller’s maximum level of workload can be influenced by many external factors such as home life, health, or the amount of quality sleep they have had. This list is not
exhaustive. This susceptibility makes it difficult to make comparisons between controllers’ abilities. Quite often, even the individual will not appreciate they might not be performing to the best of their ability.

2.60. Unexpectedly high workloads, for example, aircraft emergencies or unforecasted weather changes, could create a situation where the ATCO becomes overloaded.

2.61. Another element affecting workload is time. Andre (2001) states that workload grows with time even when the task demand is steady. Therefore, if an individual is required to move from one rating to another, the workload will increase regardless of the task demand until the ATCO is relieved from the position.

2.62. The graph below shows the variance in workload on two imaginary sectors. It is purely hypothetical, but if you imagine an ATCO that is required to move between two sectors only to meet periods of increased demand, then the ATCO could experience long periods of consistently high workload. If staffing was adequate to prevent an ATCO being moved from one rating to another, then the ATCO would not always be required to move into these high workloads consistently.

2.63. Breaks are an important way of reducing the decline in performance with increasing time on tasks due to the effects of high workload. The length of time before a break occurs, and the duration of the break should be commensurate with the type of task being performed. This is a particularly difficult limit to prescribe for the whole industry as there are many factors that can affect workload. ATCOs require breaks to sustain performance during cognitively intense periods of time-in-position. Depending on the fatigue management system in operation, either the ANSP or the State are required to provide the minimum duration of breaks between periods of time-in-position in a duty period and any significant deviations, and the reason for the deviation, should be noted (ICAO Annex 11 – Appendix 5 b) iii, Appendix 6 1.2, f).

2.64. While the State and the service provider have responsibility regarding fatigue, the individual ATCO also bears some responsibility. Individuals are responsible for making optimum use of non-work periods, coming to work fit for duty, using personal fatigue mitigation strategies while on duty and reporting fatigue issues.

Mitigations

2.65. If staffing levels are sufficient to prevent staff using both ratings on the same shift, you could reduce the risk of increased workload, risk of fatigue and therefore operational errors. Examples are South Africa and The Netherlands, both of which have approach and area in the same centre and controllers do not hold more than one rating. There is no crossover from approach to area.

2.66. The rostering system should take into consideration the expected operational duties of the individual ATCO to meet the requirement to maintain competency on all the ratings the ATCO holds.

2.67. Technical and system design should enable the reduction of mode-switching errors and operational errors. A good example would be the systemization across the technology implemented at facilities for all operational working positions. While this would go some way to reduce some of the risks involved, it is impossible to remove them all as the controlling tasks for area and approach are very different. The technology used for area could be irrelevant on approach for example. So, even with some element of standardisation in the technology used on working positions, there will always be an element of risk that comes when moving between ratings.

2.68. If controllers are to be required to move between ratings, then, education on the risks is really important. While the technical design could mitigate some risk that comes with the differences in technology, it is also important for controllers to be able to adapt their cognitive mental model and adapt their controlling methods for the new controlling position. Reaction and resolution times are likely to be different when switching between ratings and controllers should be able to prepare themselves for this. Sufficient fatigue breaks allow the controller to rest and reset themselves. Some ANSPs acknowledge this fact and only allow an ATCO to switch between ratings once during their operational shift. Another option would be the use of a checklist for the controller to enable the activation of the correct mental model in advance.

2.69. Another mitigating factor is an awareness of mode-switching errors. From the ATCO to the supervisor, people can, and do, regularly make small mistakes due to mode switching and might not either be aware, or overly concerned due to their capacity to manage such errors. Awareness allows the controller and supervisor to consider the potential risks when switching modes. Some management is seen when mode-switching risks have been identified. These can include only allowing the ATCO to move between ratings once during their operational shift, i.e., not allowing the controller to revert back to exercising the original rating.

IFATCA Technical and Professional Manual (TPM)

2.70. The TPM has no guidance for the specific scenario of using two ratings within a controller’s operational shift, however, it does contain detailed policy on certain elements of this topic. These have been referenced in this paper and they are:

• Four eyes principle WC 10.1.17;
• Duty rostering WC 10.1.7;
• Work & rest scheme WC 10.3.1;
• Recency & competency WC 10.3.4;
• Ageing ATCOs WC 10.3.7.

Conclusions

3.1. Using two ratings in a single operational shift is not particularly common, however, it does occur and there are no restrictions in place other than a requirement of adequate staffing levels, the application of time away from duty to be sufficient for fatigue breaks and the requirement for the controller to maintain their competency.

3.2. Working two ratings simultaneously could allow for the risks associated in the paper to be either exacerbated or could allow for a single mental model to be developed by the ATCO, dependent on the system and HMI design that allows the ATCO to do so. Some ANSPs already practice this method of simultaneous operation, including the provision of approach within the tower environment.

3.3. Switching between two ratings during an operational shift increases the likelihood of exposure to consistently higher workloads which in turn increases the possibility of fatigue.

3.4. While fatigue management is described within Annex 11 and ICAO Doc 9966, the overly fatiguing method of switching between two ratings is not considered specifically.

3.5. Mode-switching hazards can manifest when an ATCO moves from one rating to another. Different ratings often require the use of different technology and ATC systems, which all require different operating methods.

3.6. Aircraft and pilot behaviour varies depending on the phase of flight and may result in an ATCO expecting different closure rates, turn radii and R/T response times.

Recommendations

It is recommended that the following policy statement is moved from WC10.3.2 and placed within an entirely new section of the TPM covering the use of multiple ATCO ratings.

With the addition of:

ATCOs required to provide a surveillance approach service and a surveillance area service simultaneously should be strongly discouraged by MAs.

It is recommended that the following is also included within the new section of the TPM.

When ATCOs are required to exercise more than one rating during their operational shift, the procedure shall undergo a safety risk assessment with consideration of:

• The human performance of the ATCO, specifically the risk of mode-switching errors
• Fatigue risk management
• The ability to maintain sufficient controller competency.

References

ICAO Convention on International Civil Aviation – Annex 1, July 2022, Fourteenth Edition.

ICAO 10056 – Air Traffic Controller Competency-based Training and Assessment, Second Edition 2022, Amendment – 10th November 2022.

EASA – Easy Access Rules for Air Traffic Controllers’ Licensing and Certification (Regulation (EU) 2015/340), Published December 2019, Issue 1, Amendment 3 – 2nd Jan 2020.

IFATCA Working paper no.158 – Review of Single Person Operations Policy and Four Eyes, Principle, Bali, 2013.

IFATCA Technical and Professional Manual, version 65.

EUROCONTROL Hindsight 28th edition – Mode Switching in Air Traffic Control, February 2019.

ICAO Doc 9966 – Fatigue management guide for ATC service providers, Second Edition, Version 2 2020.

IFATCA Working paper no. 153 – Elements of the FRMS Model, Gran Canaria, 2014.

ICAO Annex 11, July 2018, Fifteenth Edition, Amendment – 5th May 2020.

Waard, D de (1996). The Measurement of Drivers’ Mental Workload.

Andre, A.D. (2001). The Value of Workload in the Design and Evaluation of Consumer Products.

IFATCA Working paper no. 158 – The Provision of Surveillance Approach and Aerodrome Control Services as Combined Function, Sofia, 2015.

IFATCA Working paper no. 161 – The Ageing Controller, Punta Cana, 2010.

IFATCA Working paper no. 303 – Cognitive Processes in Air Traffic Control, Las Vegas, 2016.