Summary

The Federal Aviation Administration (FAA) faces four challenges in identifying the level of air traffic controller staffing needed to provide safe and cost-effective services for the diverse range of aviation operations supported in American airspace. First, there are no definitive methods available to FAA or to air traffic control (ATC) providers in other countries for relating staffing to safety or airspace performance beyond historical trends. Second, FAA is working with uncertain traffic forecasts, which have often overestimated future levels of air traffic. Third, controller training requires significant lead time. Most FAA trainees require between 1½ and 3 years of on-the-job training to qualify fully for all facility positions, and even certified controllers require at least a year to recertify when they are transferred to a new facility. Fourth, the controller workforce available within the next year can be uncertain. In 2014, for example, roughly 9.8 percent of the controller workforce is predicted to be lost because of trainees failing to qualify, promotions out of controller positions, and retirements. Indeed, 3,224 controllers (21 percent of the workforce) were eligible for retirement at the end of FY 2012. Thus, FAA recently increased the number of trainees, temporarily expanding the size of the workforce relative to traffic demand in anticipation of pending retirements.

FAA must address all the challenges identified above to ensure safe and cost-effective staffing, not only at the national level but also at each of its 315 facilities. Furthermore, it must ensure that staffing continues to be appropriate as it implements the new air traffic operations environment associated with its modernization initiative, the Next Generation Air Transportation System (NextGen).

In response to long-standing debates about appropriate controller staffing, Section 608 of the FAA Modernization and Reform Act of 2012 tasked the National Academy of Sciences with studying “the air traffic controller standards used by the [FAA] to estimate staffing needs for FAA air traffic controllers to ensure the safe operation of the national airspace system in the most cost effective manner.” The term “staffing standards” is defined narrowly by FAA to mean mathematical models relating controller workload to air traffic activity (FAA 2013). These staffing standards, sometimes referred to as staffing models, are only one part of the larger process used by FAA in determining controller staffing levels. Consistent with clarification from congressional staff, the committee took a broader approach and considered the full range of processes that FAA uses in estimating the number of controllers it needs, including the input of facility managers as well as mathematical models, and the processes by which FAA ensures that controllers are properly distributed across facilities.

Overall, the committee found FAA’s staffing standards for terminal ATC facilities to be reasonable for use in developing initial estimates of the number of controllers needed for managing traffic at each facility. However, it had concerns about the validity of the mathematical model used for en route facilities and the resulting estimates of controller staffing needs. The steps taken by FAA to create a controller staffing plan from the staffing standards and then execute this staffing plan are obscure. As a result, the committee was unable to determine the extent to which staffing imbalances are being corrected over time to help ensure cost-effective staffing.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 1
Summary The Federal Aviation Administration (FAA) faces four challenges in iden- tifying the level of air traffic controller staffing needed to provide safe and cost-effective services for the diverse range of aviation operations sup- ported in American airspace. First, there are no definitive methods avail- able to FAA or to air traffic control (ATC) providers in other countries for relating staffing to safety or airspace performance beyond historical trends. Second, FAA is working with uncertain traffic forecasts, which have often overestimated future levels of air traffic. Third, controller training requires significant lead time. Most FAA trainees require between 1½ and 3 years of on-the-job training to qualify fully for all facility positions, and even certi- fied controllers require at least a year to recertify when they are transferred to a new facility. Fourth, the controller workforce available within the next year can be uncertain. In 2014, for example, roughly 9.8 percent of the controller workforce is predicted to be lost because of trainees failing to qualify, promotions out of controller positions, and retirements. Indeed, 3,224 controllers (21 percent of the workforce) were eligible for retire- ment at the end of FY 2012. Thus, FAA recently increased the number of trainees, temporarily expanding the size of the workforce relative to traffic demand in anticipation of pending retirements. FAA must address all the challenges identified above to ensure safe and cost-effective staffing, not only at the national level but also at each of its 315 facilities. Furthermore, it must ensure that staffing continues to be appropriate as it implements the new air traffic operations environ- ment associated with its modernization initiative, the Next Generation Air Transportation System (NextGen). In response to long-standing debates about appropriate controller staffing, Section 608 of the FAA Modernization and Reform Act of 2012 1

OCR for page 1
2 FAA’s Approach for Determining Future Air Traffic Controller Staffing Needs tasked the National Academy of Sciences with studying “the air traffic controller standards used by the [FAA] to estimate staffing needs for FAA air traffic controllers to ensure the safe operation of the national airspace system in the most cost effective manner.” The term “staffing standards” is defined narrowly by FAA to mean mathematical models relating controller workload to air traffic activity (FAA 2013). These staffing standards, sometimes referred to as staffing models, are only one part of the larger process used by FAA in determining controller staffing levels. Consistent with clarification from congressional staff, the committee took a broader approach and considered the full range of processes that FAA uses in estimating the number of controllers it needs, including the input of facility managers as well as mathematical models, and the processes by which FAA ensures that controllers are properly distributed across facilities. Overall, the committee found FAA’s staffing standards for terminal ATC facilities to be reasonable for use in developing initial estimates of the number of controllers needed for managing traffic at each facility. However, it had concerns about the validity of the mathematical model used for en route facilities and the resulting estimates of controller staff- ing needs. The steps taken by FAA to create a controller staffing plan from the staffing standards and then execute this staffing plan are obscure. As a result, the committee was unable to determine the extent to which staff- ing imbalances are being corrected over time to help ensure cost-effective staffing. SAFETY IN STAFFING ATC is considered vital to the safety of aviation operations. However, the relationship between controller staffing levels and aviation safety is not well understood. Various FAA organizations gather data related to safety; examples include data on incidents violating various safety criteria without causing accidents, records of actual operations, and voluntary controller reports from the Air Traffic Safety Action Program. However, the committee found no systematic and proactive mechanisms within FAA for (a) analyzing these data for concerns relative to staffing levels or (b) involving the controller workforce in discussions about staffing con-

OCR for page 1
Summary 3 cerns as knowledge about relevant safety issues emerges. Thus, FAA does not have the data to anticipate with any certainty the safety effects of changes in current controller staffing levels or changes in air traffic opera- tions with NextGen. Recommendation 1 FAA should explore the relationships between controller staffing and safety by • Analyzing the wide range of data that can identify relationships between staffing and safety, including accident and incident reports, voluntary reports by controllers from the Air Traffic Safety Action Program, and other databases that, if properly integrated, can relate safety to staffing concerns (e.g., records of actual shifts worked); and • Involving the controller workforce in staffing decisions, particularly as knowledge concerning relevant safety issues emerges. FAA should use insights gained from these activities to inform deci- sions about controller staffing levels associated with the transition to NextGen and any other policies likely to result in changes in historically safe staffing levels. {2-5, 5-5}1 DETERMINATION OF WORKFORCE SIZE The size of the controller workforce is based on three general steps: (a) point estimates derived from models, including forecasts of air traffic demand; (b) expansion of point estimates into ranges that incorporate input from facility managers; and (c) a hiring plan and transfer pro- cess that result in net changes to the total workforce and its distribution across FAA’s 315 facilities. The first step, point estimates for each of the facilities, is the output of the on-position staffing models. The models predict the number of controllers needed on position to perform traffic-related tasks, most notably, separating aircraft from one another. However, the models do 1 The numbers in braces following the summary recommendations refer to related recommenda- tions in the report chapters.

OCR for page 1
4 FAA’s Approach for Determining Future Air Traffic Controller Staffing Needs not explicitly account for a number of important off-position controller tasks, such as ongoing training to maintain certification, provision of training for new controllers, participation in safety initiatives, and sup- port of new technology development and implementation. In the committee’s judgment, the models used to estimate control- ler staffing requirements for airport control towers and terminal radar approach control facilities are mostly reasonable for their purpose, with the exception of the current scheduling algorithm. However, in the case of the task load model for en route facilities, the committee shares the concerns of a predecessor committee, which focused exclusively on this model (TRB 2010). The 2010 report recommended actions to improve the model, but FAA2 has taken only limited steps toward implementing these recommendations. Thus, the current committee cannot assess with confidence whether FAA’s staffing model for en route facilities is on track to meet the recommendations of the 2010 report. The current commit- tee had a broader charter than the authors of the 2010 report: to examine the entire staffing process as opposed to only the en route staffing model. In this context, the current committee questions whether the detailed task load model is appropriate for staff planning. The model’s level of detail adds considerably to the cost and difficulty both of validating all of its parameters and of updating the model to describe the new types of operations envisioned under NextGen. Because FAA “staffs to traffic,” its staffing estimates are scaled by fore- cast changes in aviation traffic. FAA’s traffic forecasts, at least since 2000, have consistently overestimated future levels of air traffic. The models applied by FAA are not sufficiently documented to explain or justify the overestimates. In practice, a high estimate one year can be followed by lower hiring the next year from a national perspective, but overstaffing can be created at individual facilities that do not experience enough attri- tion to correct any imbalance. Thus, overly optimistic traffic forecasts can have a lasting impact. In the second step, the model-based staffing standards are combined with productivity data, where available, and with assessments of staffing 2 The task load model for en route facilities was developed by MITRE Corporation under contract to FAA.

OCR for page 1
Summary 5 needs from operational managers to generate a desired staffing range for each facility. These ranges are reported to Congress annually in FAA’s controller workforce plan. The input from operational managers, known as service unit input (SUI), is intended to reflect each facility’s unique operational requirements. FAA’s ongoing efforts to establish consistent methods for capturing SUI appear to address issues raised in an earlier report (TRB 1997). However, these methods have not been clearly docu- mented and can appear arbitrary. For the third step, the hiring of new controllers and the transfer of current controllers from one facility to another are FAA’s primary mecha- nisms for adjusting staffing levels. The annual hiring plan for each facility is developed by FAA’s Office of Labor Analysis and its Air Traffic Organi- zation on the basis of staffing targets (derived from the staffing ranges) and operational and training constraints. As in the case of other elements of staff planning, this negotiation is not fully documented and can appear subjective. Requests for transfer are initiated by controllers themselves, and FAA management makes no attempt to encourage controllers to move from facilities staffed above their target to facilities staffed below their target. Not surprisingly, therefore, the committee finds that transfers appear to be poorly coordinated and do not achieve their potential of redistributing the workforce to meet facility targets. Taken in its entirety, the staffing process by which FAA determines the total number of controllers can sometimes appear arbitrary, both to this committee and to the organizations and workforce that need to imple- ment the staffing plan within FAA. This concern arises because staff planning is not consistently documented and because it can be modi- fied by various organizations within FAA in uncoordinated ways. The model-based staffing standards themselves are clearly documented, but this is not the case for other parts of the process. In the committee’s judgment, justification and consistent documentation and application of the methods used to determine the size of the controller workforce are critical for informed, data-driven decision making about staffing needs and hiring. Furthermore, the overall cost-effectiveness of FAA’s controller staffing process depends not only on developing a robust staff- ing plan but also on FAA’s ability to train and position these controllers appropriately at specific facilities. The body of the report offers several

OCR for page 1
6 FAA’s Approach for Determining Future Air Traffic Controller Staffing Needs suggestions for enhancing the effectiveness of the training, hiring, and transfer processes. Recommendation 2 FAA should reassess its approach to developing an improved staffing model for en route facilities and make any necessary changes, poten- tially including the adaptation or formulation of a new model likely to be developed and validated in a timely manner and at reasonable cost. Any new model should be constructed in such a way that it can be updated as NextGen operations are implemented. {3-2} Recommendation 3 FAA should take steps to ensure that the planning and execution of its air traffic controller staffing process are clear, consistent, and transpar- ent to a range of stakeholders. Stakeholders include but are not limited to the following: • The controller workforce, which needs to engage with FAA in the col- laborative development of improved staffing plans and their execu- tion to ensure overall cost-effectiveness; and • Congress, which needs to make informed decisions about future budgets for controller staffing. {3-3, 4-2, 4-3} COST-EFFECTIVE AND SAFE SCHEDULING Work schedules determine how many controllers report to a facility at any given time, when they take breaks, and how long a recuperative period they have between work shifts. Schedules can affect whether controller staff are used in a cost-effective manner, particularly at larger facilities, which can benefit from economies of scale. In addition, scheduling can affect safety. Extensive evidence shows fatigue to be a risk factor in 24/7 operations such as ATC facilities. Rare but widely publicized incidents of FAA controllers falling asleep on the job have highlighted the issue. FAA has begun establishing a Fatigue Risk Management System, includ- ing a working group involving controllers, management, and experts in fatigue. However, under recent budget cuts, FAA has effectively eliminated

OCR for page 1
Summary 7 the Fatigue Risk Management System program’s capability of monitor- ing for fatigue concerns proactively and of investigating whether recent initiatives to reduce fatigue risks are providing the intended benefits. Air navigation service providers in other nations, including Canada, Germany, and Australia, have implemented new scheduling software. FAA uses only a simple scheduling algorithm at a national level in gen- erating staffing standards. A major limitation of this algorithm is its inability to schedule shifts that start one day and end the next (i.e., that cross midnight). Given concerns about controller fatigue on midnight shifts, this limitation is particularly problematic, and the algorithm may generate staffing levels insufficient for adequate fatigue mitigation. In addition, FAA headquarters provides no consistent guidance or tools to local facilities to help them develop their operational schedules. As a result, each facility develops its own schedule independently of FAA’s staff planning process. The actual controller schedules may not reflect key assumptions in the staff planning, may not be the most cost-effective, and may not incorporate best practices in fatigue risk management. FAA is contracting with the same vendor used by air navigation ser- vice providers in other countries to implement a new scheduling tool, but the timeline of its implementation at all facilities is not fixed.3 The following are among the potential benefits of sophisticated scheduling software: • It would provide a consistent basis for establishing work schedules that minimize or mitigate the safety risks associated with controller fatigue. • It would ensure that diverse facilities are all capable of generating effi- cient schedules, particularly at larger facilities where economies of scale may be possible. • It would provide a consistent basis for informing the development of staffing standards at FAA headquarters and the creation of work schedules at the facility level. Schedule changes significantly affect the controller workforce. FAA will, therefore, need to collaborate closely with the National Air Traffic 3 FAA’s target date for implementing the new scheduling tool at 15 facilities (the end of FY 2013) appears to have slipped.

OCR for page 1
8 FAA’s Approach for Determining Future Air Traffic Controller Staffing Needs Controllers Association in implementing an improved scheduling capa- bility and in adopting revised schedules that address fatigue. Recommendation 4 FAA should, as a matter of priority, continue its efforts to develop an improved scheduling tool capable of creating efficient controller work schedules that incorporate fatigue mitigation strategies. The agency should collaborate closely with the National Air Traffic Controllers Association in implementing this improved scheduling capability, notably in adopting schedules that reflect science-based strategies for managing the risks associated with controller fatigue. {2-3, 3-1, 4-9} BUDGETS AND COST-EFFECTIVENESS The cost of FAA’s workforce of about 14,900 controllers in FY 2014 is estimated at $2.8 billion—about 18 percent of the total FAA budget and 29 percent of the Operations budget. FAA’s capital expenses and some of its operating expenses are covered by the Airport and Airway Trust Fund. Over the past decade or more, the FAA Operations budget has required a substantial and growing amount of support from the General Fund— $4.4 billion in 2013, including $1.3 billion that helped to pay for the ATC workforce. Congressional concerns about the cost-effectiveness of FAA’s staffing models are driven, in part, by the growing cost of the ATC staff and the growing reliance of the FAA Operations budget on general revenues over the past decade. In a time of fiscal austerity and stalemate in Congress with regard to deficit budgets and taxes, continuing to depend on the General Fund makes FAA Operations vulnerable to budget cuts. Revised forecasts of rebounding aviation trust fund receipts by the Congressio- nal Budget Office (CBO) in 2013 and the administration in 2014 imply that growing trust fund revenues will be able to reverse the demand for General Fund revenues. In contrast to the $4.4 billion of General Fund revenues received in FY 2013, the administration is requesting only $700 million for FY 2015. Whether the administration’s and CBO’s revised forecasts will hold is open to question. Both forecasts imply faster

OCR for page 1
Summary 9 growth in aviation demand than in gross domestic product or in FAA’s projected rate of increase in revenue passenger miles. In case revised forecasts of aviation trust fund receipts prove optimis- tic and to help put concerns with regard to the cost of the ATC work- force in perspective, an illustrative example of a cut in ATC staffing of 8 percent is described in Chapter 6. Such a cut would reduce the cost of the ATC workforce by about $223 million, which represents about 1.4 percent of FAA’s annual budget and 7 percent of the General Fund revenues used for FAA Operations in 2014. Such a cut in staffing would have unknown effects on aviation safety and service and should not, therefore, be taken as a suggestion. Air traffic has declined significantly since its peak in 2000 and is not expected to return to that level in the near term (FAA 2013). Meanwhile, controller staffing levels are similar to those in 2000. However, the systemwide data do not indicate that all ATC facilities are overstaffed or that controller productivity has dropped dramatically in all facilities over the past 13 years. Indeed, detailed recommendations within the report address some important facilities that appear to be chronically under- staffed. Comparisons of controller staffing over time are complicated by changes in the composition of the workforce—for example, the high percentage of controllers eligible to retire and the large number of train- ees being brought in to replace them. Furthermore, the volume and nature of traffic vary significantly among ATC facility types. For example, while almost all types of opera- tions have been reduced since 2000, the decline in ATC operations has been particularly pronounced at smaller towers. Thus, broad general- izations about controller productivity mask important variations at the level of individual facilities. At some larger facilities, changes in staffing levels may affect the throughput of busy airspace and could result in direct costs or benefits to aircraft operators. In contrast, at smaller facili- ties, staffing levels may depend on minimum staffing requirements that are not driven by traffic levels but instead by the hours when the facility is required to provide service. Defining the most cost-effective staffing model, as requested in the committee’s charge, requires safety and performance metrics that FAA has not defined or assessed. Thus, the committee’s recommendations presented in this summary and in the report aim to enable controller

OCR for page 1
10 FAA’s Approach for Determining Future Air Traffic Controller Staffing Needs staffing decisions that are consistent; that are driven by proper science and data analysis; and that will address relationships between ensuring safety, meeting the operational needs of the aviation community, and demonstrating cost-effectiveness. REFERENCES Abbreviations FAA Federal Aviation Administration TRB Transportation Research Board FAA. 2013. A Plan for the Future: 10-Year Strategy for the Air Traffic Control Workforce, 2013–2022. http://www.faa.gov/air_traffic/publications/controller_staffing/media/ CWP_2013.pdf. TRB. 1997. Special Report 250: Air Traffic Control Facilities: Improving Methods to Deter- mine Staffing Requirements. TRB, National Research Council, Washington, D.C. TRB. 2010. Special Report 301: Air Traffic Controller Staffing in the En Route Domain: A Review of the Federal Aviation Administration’s Task Load Model. Transportation Research Board of the National Academies, Washington, D.C.