4

Development and Implementation of Staffing Plan

The model-based staffing standards and the staffing ranges described in Chapter 3 are developed by the Federal Aviation Administration’s (FAA’s) Office of Labor Analysis (ALA) to inform an annual hiring plan. The goal of the hiring plan is to move each facility to a level consistent with the staffing range. The Air Traffic Organization (ATO) and ALA then develop an annual staffing plan, including new hires and transfers among existing staff, that serves as the basis for decisions concerning each of FAA’s 315 operational air traffic control (ATC) facilities. The execution of the staffing plans, which results in new hires and transfers and in the placement of personnel among facilities, is carried out by ATO.

This chapter focuses on the methods applied by FAA in its detailed staff planning and execution. In some cases the methods were unclear to the committee; they are analyzed on the basis of FAA data representing typical inputs to the methods and their outputs. The efficacy of staff planning and execution depends on the extent to which the result is consistent with the published controller workforce plan. That measure can change as FAA’s methods change.

The chapter describes how FAA assesses current facility staffing status and explores how FAA develops its hiring and staffing plans. It describes the committee’s understanding of the execution process and illustrates how the number and location of controllers in the workforce compare with the goals of the staffing plan. Potential strategies for improvements in staff planning are noted. The final section summarizes the chapter and gives the committee’s recommendations.

FACILITY STAFFING STATUS

Assessment of Status

Staff planning requires a target and a clear path for reaching it. In concept, plans for individual facilities span at least 3 years to account for losses due to attrition (such as retirements, promotions, and deaths) and the period needed to select, hire, and train personnel to reach full certification, which typically ranges from 1 to 3 years. Achieving the target staffing level over time for a facility requires a good understanding of the current staffing status. ALA uses the head count of certified professional controllers (CPCs) and certified professional controllers in training (CPC-ITs) to assess each facility’s staffing status relative to the staffing range described in Chapter 3. Status is assessed as follows:

• Above range: CPC + CPC-IT is more than 10 percent above the staffing range midpoint.

• Within range: CPC + CPC-IT is within ±10 percent of the staffing range midpoint.

• Below range: CPC + CPC-IT is more than 10 percent below the staffing range midpoint.



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 53
4 Development and Implementation of Staffing Plan T he model-based staffing standards and the staffing ranges described in Chapter 3 are developed by the Federal Aviation Administration’s (FAA’s) Office of Labor Analysis (ALA) to inform an annual hiring plan. The goal of the hiring plan is to move each facility to a level consistent with the staffing range. The Air Traffic Organization (ATO) and ALA then develop an annual staffing plan, including new hires and transfers among existing staff, that serves as the basis for decisions concerning each of FAA’s 315 operational air traffic control (ATC) facilities. The execution of the staffing plans, which results in new hires and transfers and in the placement of personnel among facilities, is carried out by ATO. This chapter focuses on the methods applied by FAA in its detailed staff planning and execution. In some cases the methods were unclear to the committee; they are analyzed on the basis of FAA data representing typical inputs to the methods and their outputs. The efficacy of staff planning and execution depends on the extent to which the result is consistent with the published controller workforce plan. That measure can change as FAA’s methods change. The chapter describes how FAA assesses current facility staffing status and explores how FAA develops its hiring and staffing plans. It describes the committee’s understanding of the execution process and illustrates how the number and location of controllers in the workforce compare with the goals of the staffing plan. Potential strategies for improvements in staff planning are noted. The final section summarizes the chapter and gives the committee’s recommendations. FACILITY STAFFING STATUS Assessment of Status Staff planning requires a target and a clear path for reaching it. In concept, plans for individual facilities span at least 3 years to account for losses due to attrition (such as retirements, promotions, and deaths) and the period needed to select, hire, and train personnel to reach full certification, which typically ranges from 1 to 3 years. Achieving the target staffing level over time for a facility requires a good understanding of the current staffing status. ALA uses the head count of certified professional controllers (CPCs) and certified professional controllers in training (CPC-ITs) to assess each facility’s staffing status relative to the staffing range described in Chapter 3. Status is assessed as follows: • Above range: CPC + CPC-IT is more than 10 percent above the staffing range midpoint. • Within range: CPC + CPC-IT is within ±10 percent of the staffing range midpoint. • Below range: CPC + CPC-IT is more than 10 percent below the staffing range midpoint. 53

OCR for page 53
54 Federal Aviatio Administrat F on tion’s Approac for Determin ch ning Future Ai Traffic Contr ir roller Staffing Needs In FAA’s terminolog the “abov range” cat s gy, ve tegory is ref ferred to as “ “high” staffin the “with ng, hin range” caategory as “a adequate” staffing, and the “below r ange” catego as “low” staffing. t ory ” Figure 4- is a modif version of a figure th appears i the 2013 controller w -1 fied hat in workforce pla It an. shows seeveral impliccations and potential cau of above and below p uses e- w-range staff fing. At the end of FY 2012, of FAA’s 315 ATC facili A f f 5 ities, 135 (re epresenting 5 percent of the 53 f CPC + CPC-IT work C kforce) were classified as above rang 102 (repr s ge, resenting 34 percent of th he CPC + CPC-IT work C kforce) were classified as within rang and 78 (r s ge, representing 13 percent o g of the CPC + CPC-IT workforce) were classifie as below r w w ed range (Table 4-1). Acco e ording to FAAA, the large proportion of facilities that are abov range is e o t ve explained in part by the ggradual FIGURE 4-1 Illustr E ration of controller staffing range. .

OCR for page 53
Development and Implementation of Staffing Plan 55 TABLE 4-1 Certified Controller Staffing Levels of FAA Facilities (End FY 2012) Facilities CPC + CPC-IT Head Count Facility Staffing Level Number Percent Number Percent Above range 135 43 6,855 53 Within range 102 32 4,319 34 Below range 78 25 1,722 13 Total 315 100 12,896 100 movement of the retirement bubble through the workforce. The peak of controller retirements occurred in 2007, but as of FY 2012, about 3,000 controllers were still eligible to retire (i.e., roughly 20 percent). Historical trends indicate that most of this population will not retire until their 7th year of eligibility (FAA 2013, 31), and between 606 and 742 CPCs are expected to retire each year between 2013 and 2017 (FAA 2013, 33). As the wave of retirements continues, FAA expects that many facilities currently above range will fall within range. Facilities that are below, within, or above the staffing range are of all types—smaller towers that mostly serve general aviation; towers that serve small and medium airports with both general aviation and commercial traffic; and the largest airport towers, terminal radar approach control (TRACON) facilities, and centers (Table 4-2). Facilities with relatively few staff (12 or fewer CPCs) are particularly sensitive to sudden small changes in numbers of personnel that would change their classification. A large proportion of facilities below the staffing range are towers serving mostly general aviation and small and medium-sized airports. Nevertheless, some facilities serving important, high-volume airports and airspace fall into this category. A few of these facilities persistently appear in this category and have been deemed as chronically “hard to staff” for a variety of reasons. For example, according to the FY 2013 controller workforce plan, the staffing range for the Oakland, California, en route center is 185 to 226, whereas it has only 154 CPCs. The staffing range of the New York TRACON (serving JFK, LaGuardia, Newark, and Teterboro Airports) is 178 to 218, whereas it has 156 CPCs.1 Each of these facilities is assigned CPC-ITs and developmental controllers to raise its total staffing level to at least the bottom of the range. However, new personnel are not qualified to staff all the positions at the facilities, and current CPCs must spend time training them. San Juan and Guam, which pose their own unique challenges, are included among the hard-to-staff facilities. TABLE 4-2 Facility Staffing Relative to Staffing Range, FY 2012 Below Within Above Major Facility Type N % N % N % Smallest towers, mostly for general aviation 21 27 16 16 36 27 Small and medium-sized airport towers 45 58 56 55 56 41 Large towers, TRACONs, and en route centers 12 15 30 29 43 32 Total 78 100 102 100 135 100 1 FAA calculates a facility’s staffing status on the basis of combined CPCs and CPC-ITs. In FY 2013, this metric indicated that both the Oakland center and the New York TRACON were below range.

OCR for page 53
56 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs Many staff at facilities that are classified as above the range are located at air route traffic control centers (en route centers). The 23 en route centers have 200 to 400 certified controllers each and represent a large proportion of the fully certified controllers in the system (roughly 44 percent at the end of FY 2012). About half of the centers (12) were in the “above” category at the end of FY 2012. Roughly 30 percent of the CPCs in the centers classified as above the range midpoint were eligible for retirement in FY 2012. These centers tend to have large numbers of CPC-ITs and developmental controllers receiving training to replace CPCs as they retire. Moving a facility that is classified as above or below the range toward its staffing target is a multiyear process. Each facility’s workforce may contain “homesteader” controllers who do not wish to move once they settle into a facility, establish households, and build ties to their communities. Inducing staff to move away from facilities that are within or above the staffing range to those that are below the range can be difficult. Building up a facility that is understaffed requires lead time for bringing new personnel on board (either from other facilities or as new hires) and training them. For both new and transferring employees, the training time varies dramatically with the individual’s level of experience. Data provided by ALA allowed the committee to analyze a variety of additional metrics relative to the above–within–below classifications in the FY 2013 controller workforce plan. The metrics examine workforce composition, overtime usage, and time on position from the perspectives of “working certified positions” and “on-the-job training.” Table 4-3 shows how the metrics vary with staffing status. The following are observations on the metrics presented in the table: • On the basis of the CPC + CPC-IT assessment, the “below” and “above” staffing level percentages are more than 10 percent away from the staffing range boundaries—the group above the range midpoint is roughly 20 percent above and the group below the range midpoint is almost 20 percent below.2 • The controller pipeline (all staffing qualification levels except CPC) constitutes 23 percent of the workforce. On the basis of an analysis of ALA-provided data, that number has remained little changed, ranging from 23 to 27 percent during the preceding 5 years. • As would be expected, overtime usage at sites below the staffing range is more than twice that of sites above the staffing range. • As the staffing level relative to the staffing range midpoint increases, CPCs perform a greater portion of the total time on position. This implies that in facilities categorized as below range, CPC-ITs and developmental controllers contribute a larger share of productive work because the facilities lack an adequate number of fully qualified CPCs. • The proportion of CPC time on position devoted to on-the-job training increases slightly at the facilities categorized as below range. This is likely because CPCs at facilities that are not adequately staffed need to provide developmental controllers with more training time. 2 These estimates are based on the numbers of staff in each category (the total staff in all facilities that are below, within, or above range compared with the total staff estimated by the staffing range midpoint). The estimates are not based on the average of each facility’s rating relative to the staffing range midpoint, since the estimate would be skewed by the small number of small facilities that are substantially below or above the staffing range.

OCR for page 53
Development and Implementation of Staffing Plan 57 TABLE 4-3 Metrics for FAA ATC Facilities Above Within Below Systemwide Range Range Range Staffing level (CPC + CPC-IT) as percentage of 120 100 81 106 staffing range midpoint Percentage of staff that are CPCs 83 74 83 77 Overtime hours as percentage of all hours 1.4 3.5 3.6 2.4 worked Percentage of time on position performed by 90 85 80 87 CPCs Percentage of CPC time on position also 8 10 11 9 providing on-the-job training to non-CPCs Percentage of non-CPC time on position also 41 35 30 36 receiving on-the-job training NOTE: Non-CPCs = certified professional controllers in training + developmental controllers. Are There Other Methods for Comparing Current Staffing Levels with Staffing Ranges? The staffing standards process described in Chapter 3 estimates the number of controller positions that need to be staffed during every 15-minute interval and develops nominal schedules, daily staffing, and overall facility staffing strengths to meet these demands. No consideration is given to the extent to which each facility will be staffed by CPCs, CPC-ITs, and developmentals. According to FY 2012 data provided to the committee by ALA, CPCs made up only 77 percent of the workforce; the remainder was made up of CPC-ITs (8 percent), developmentals (10 percent), and recent hires at earlier stages (FAA Academy graduates at the operational sites, 4 percent; candidates at the academy, 1 percent). In FY 2012, some 11,753 CPCs served 10.6 million hours on position (averaging 902 hours per year per controller); the 3,310 CPC-ITs, developmentals, and academy graduates served 1.6 million hours on position (averaging 496 hours per year). This limited comparison suggests that the average non-CPC is “working certified positions” at a rate slightly more than half (55 percent) that of a CPC and that the total non-CPC workforce performs 13 percent of the required time on position. No clear distinction between the contributions of CPC-ITs and developmental controllers can be made. CPC-ITs typically progress more quickly than developmentals, but many factors can cause differences among facilities. For example, a CPC who transfers from a small tower to an en route center will need to undergo the same extensive training as a new hire would experience and may have a similar chance of successful certification. Thus, how should actual staffing levels be compared with staffing ranges that were generated, in large part, from the staffing standards? The method used in the controller workforce plan assesses the sum of CPCs and CPC-ITs at each facility, even though the CPC-ITs may not be able to staff each position. Furthermore, carving out CPC-ITs in the baseline assessment of facility staffing status in such a generalized form fails to account for nuances in training progression mentioned above and could result in an inaccurate depiction of facility staffing capacity. A more conservative method would compare the CPCs at each facility with the staffing range. By this method, 113 of the 315 facilities fall below the staffing range minimum and 62 exceed the maximum. However, this method does not consider the impact of the CPC-ITs and developmentals at each facility. This omission can have the following effects:

OCR for page 53
58 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs • The progress of the CPC-ITs and developmentals through the certification process and the contributions of the CPC-ITs and developmentals in terms of the positions that they are able to staff would not be reflected. A comparison of the number of CPCs at a facility with its staffing range would be misleading in a short-term assessment of facility staffing. • The additional training demands created by the CPC-ITs and developmentals on the facility, notably the on-the-job training provided by the CPCs, would not be reflected. Thus, a comparison of the number of CPCs at a facility with its staffing range may incorrectly imply that it is within or above range. • The extent to which the comparison of the number of CPCs with the staffing range may change once staff qualify at a facility would not be reflected. A facility that appears to be below range in one year may appear to be within or above range the next, not because its staffing levels have changed but because of successful training of the staff already at the facility. An assessment based on a more detailed analysis of the certification status of the personnel at a facility would be more relevant. Attrition models could be refined to reflect historical trends in achieving various position certifications at a given type and level of facility (or even at an individual facility). A controller equivalent workforce (CEW) metric could emerge as a measure of a facility’s current status given the qualification levels of its controllers. Such a metric would allow for within-year comparisons of staffing levels with the staffing ranges and provide a perspective with regard to planning for hires. Such a comparison could help flag facilities that are lagging in reaching necessary levels. A CEW metric will require careful analysis and development. Its application should be limited to assessing the capacity of a facility to staff its positions and bring its CPC-ITs and developmentals up to fully qualified status. That capacity may vary with several factors, including facility type and the ratio of CPCs to trainees (i.e., CPC-ITs and developmentals). The notional estimates of relative contributions given above may serve in the aggregate, but the CEW of a facility should be calculated in a manner that reflects the facility’s capacity to staff its positions and train its new members.3 For example, a facility with a large number of CPC-ITs and developmentals may have difficulty in scheduling them into positions for which they are qualified, which would limit their effective contribution. Furthermore, the CPCs may not be able to staff positions as assumed in the staffing standards while providing on-the-job training (i.e., the CEW of such CPCs may need to be less than 1.0). (Such analysis may substantiate the general axiom within FAA that CPCs must make up at least 65 percent of a facility’s workforce.) In contrast, at a facility with a high proportion of CPCs, it may be easier to schedule CPC-ITs and developmentals into positions where they can contribute and easier for CPCs to provide them with on-the-job training. 3 The estimates of time on position by controllers with varied levels of certification discussed in this section are based on data provided by ALA. If the CEW approach to estimating contributions to workload by controllers in training is pursued by FAA, it would need to verify that time on position data are captured accurately.

OCR for page 53
Developme and Implem ent mentation of St taffing Plan 59 STAFF PLANNING P G The controller workfforce at indiv vidual facilit is in con ties nstant flux. T There is a rec curring need to d reevaluat the staffin status at each facility and identify the strength adjustment required fo te ng a y h ts or maintaining or achievving optimal staffing lev l vels. Figure 4 provides an overview of the pro 4-2 s w ocess. FIGURE 4-2 Contr E roller workfforce planni process from staff p ing planning th hrough achievem ment of staff fing level.

OCR for page 53
60 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs The annual hiring plan is an important strength management tool. The process begins with ALA. For each site, the general approach consists of the following: • Projecting the population forward to a specified date, with retirements, promotions, developmental failures, and other attrition factored in; • Identifying the gap between the result of this projection and the staffing target (be it staffing standard or staffing range midpoint); and • Estimating the number of new hires needed to close the gap. Let y denote the year for which the number of new hires is being estimated. For each facility f, define the following: • t(f) = length of the training cycle for facility f. This is the nominal time required for a new hire to achieve full qualification (CPC status) at facility f. ALA uses 1-, 2-, and 3-year values in its planning. Then y + t(f) is the year in which controllers hired in year y would achieve CPC status at facility f. • p(y + t(f)) = number of CPCs projected to be at facility f in year y + t(f). The number includes residual CPCs from the beginning of year y and the various non-CPCs who achieve CPC status during the interval. ALA uses historical attrition and developmental data to estimate the surviving population in both categories. • tgt(y + t(f)) = staffing target for facility f in year y + t(f).4 • a(t(f)) = attrition rate of new hires at facility f over the training cycle. The survival rate of new hires over that interval is thus 1 – a(t(f)). • h(y, f) = number of new hires planned for facility f in year y. If p(y + t(f)) ≥ tgt(y + t(f)), h(y, f) = 0. If p(y + t(f)) ˂ tgt(y + t(f)), h(y, f) = [tgt(y + t(f)) – p(y + t(f))]/[1 – a(t(f))]. These formulas indicate that, in developing a hiring plan for a facility, ALA takes into account the status of the facility’s workforce and how it is expected to change over time, the training cycle appropriate for the facility, and the number of new hires that will be required to meet the staffing target. ANNUAL STAFFING PLAN After the hiring plan is complete, ALA gathers ATO’s input into what will eventually be an agreed-on staffing plan that will be submitted to the field for execution. The staffing plan consists of a hiring plan (for new hires) and a transfer plan (for current controllers moving between facilities). The annual staffing plan should provide the targets to make the necessary strength adjustments so that each facility can achieve a staffing level consistent with its staffing target. The staffing target in the past has been the number of staff required to meet the projected traffic that is generated by the staffing models (i.e., the staffing standard described in Chapter 3). 4 Before the FY 2014 controller workforce plan, the staffing target was the staffing standard. For the FY 2014 controller workforce plan, both the staffing standard and the staffing range midpoint will be evaluated as the staffing target.

OCR for page 53
Development and Implementation of Staffing Plan 61 For FY 2014, the committee was informed that the target could be either the staffing standard or the midpoint of the staffing range. After discussion with ATO staff, the committee could not fully discern the criteria used in determining the staffing target, other than that ATO takes into account unique circumstances that the staffing models do not represent. According to ALA, ATO staff offer suggestions as to which facilities should be the recipients of transfers (“transfers in”) but do not offer the “transfers out” detail. Since transfers in and transfers out must net to zero systemwide, ALA uses historical data to estimate which facilities are traditionally donors of transfers (i.e., transfers out of those facilities). No out–in connection is directly identifiable, but this procedure does generate a planned net change (transfers in – transfers out) at each facility. A transfer action may require two training cycle events—one for the previously qualified controller at the new facility and one for the replacement controller at the previous site. Table 4-4 provides a systemwide comparison of the ALA hiring plan with the ALA–ATO agreed-on plan with the above-, within-, and below-range facility staffing level classifications.5 The committee was surprised by the sizable increase in new hires (59 or 6.5 percent) in the staffing plan over the 902 proposed in ALA’s plan. The reasons for the difference are not fully understood. ATO staff indicated that the difference was due to adjustments to account for real- world conditions at facilities but could provide no documentation concerning how the adjustments are made or the criteria used in making them. (A subsequent section of this chapter addresses the need for more transparency in the process by which facilities provide input into the staffing plan.) In addition, the transfer process does not appear to be used proactively to steer facilities toward their staffing targets, as is described in more detail below. TABLE 4-4 Comparison of FY 2012 ALA Hiring Plan and ALA–ATO Staffing Plan ALA–ATO Staffing Plan Facility Staffing Number ALA Hiring Hiresa Transfers Inb Transfers Out Level of Sites Plan Quantity Above range 135 547 383 252 171 Within range 102 284 373 198 200 Below range 78 71 205 113 192 Total 315 902 961 563 563 a The ALA–ATO staffing plan is considered the combined output of the ALA–ATO agreed-on hiring plan and the net of (transfers in – transfers out). b Although no transfers are planned in the en route environment, many occur. [In addition to the CPC-IT transfer plan (within terminal facilities only), ALA’s hiring models include an assumed net transfer plan between en route and terminal facilities.] 5 The data for FY 2013 are available, but the effects of the sequester truncated hiring at midyear and would make the analysis unbalanced.

OCR for page 53
62 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs STAFFING PLAN EXECUTION Even the best staffing plan will not achieve the desired outcomes if it is not executed as planned. To gain a better understanding of the relationship between the staffing plan and execution of the plan, the committee compared planned staff additions and transfers in FY 2012 with what actually occurred. The committee chose FY 2012 rather than FY 2013 for the comparison because the Budget Control Act of 2011 prevented FAA from hiring new staff during a significant part of 2013. (The Budget Control Act of 2011 and its impact on FAA’s budget are discussed in Chapter 6.) The committee examined planned and executed staffing gains at each facility (Table 4-5). For purposes of this table, gains are defined as new hires plus all transfers into a facility less all transfers out. The gains are divided between terminal (tower and TRACON) facilities and en route centers and between facilities that were below, within, and above the staffing ranges (the ALA metric of CPC + CPC-IT was used as the determinant of facility staffing). Table 4-5 indicates that FAA is executing its hiring and transfer plans as intended in the aggregate. Terminals that fall below the staffing range did not achieve the intended target, but the difference between the target and the number achieved is small (five), so the discrepancy may not be difficult to overcome. Among facilities that fall above the staffing range, the category for which underperformance is appropriate, FAA met 91 percent of the target. Facilities that fall within the range slightly overachieved, but the difference (nine) is small relative to the total workforce. The impact of staffing is felt at the facility level. In examining facility-level data, the committee noticed cases from FY 2011 through FY 2013 where execution did not meet the plan. The following are examples: • The Denver (Colorado) TRACON (D01), which is staffed within the staffing range, had an aggregate (3-year) staffing plan (hires plus net transfers) of 25; it executed 15. • The Dallas–Fort Worth (Texas) tower, which is staffed above the staffing range, executed 10 against a plan of seven in FY 2013 because transfers in exceeded transfers out. • The Detroit (Michigan) TRACON (D21), which is staffed below the staffing range, executed three against a plan of 13. Transfers went fairly well according to plan, but no new hires were executed against a plan of seven. • The High Desert (California) TRACON (E10), which is below the staffing range, executed 11 against a plan of five. The overexecution was due to seven new hires in FY 2013 against a plan of one. • Guam (ZUA), also below the staffing range, executed seven new hires against a plan of nine. No transfers in or out were planned; actual transfers out of seven exceeded actual transfers in of four. The result was an underexecution by five. Examining staffing plans for any single year in terms of whether a facility is below, within, or above the staffing range could be misleading. As noted above, facility staffing is constantly changing because of retirements, other losses (deaths and attrition), new hires, transfers, and changes in staff capability as developmentals and CPC-ITs achieve CPC status. However, the FY 2012 staffing plan allocates staff to facilities in rough proportion to their current staffing levels, regardless of whether they are classified as below, within, or above the

OCR for page 53
Development and Implementation of Staffing Plan 63 TABLE 4-5 Planned Versus Executed Staffing Gains in FY 2012 Facility Staffing Status Relative to Range Below Within Above Terminals Planned gains 84 166 170 Executed gains 79 169 148 Executed/planned 94% 102% 87% Centers Planned gains 42 205 294 Executed gains 43 211 275 Executed/planned 102% 103% 94% Total Planned gains 126 371 464 Executed gains 122 380 423 Executed/planned 97% 102% 91% range. Facilities that are below the range represented 13 percent of total staff at the beginning of FY 2012 and were to receive 13 percent of staffing gains according to the plan, facilities that are within the range represented 35 percent of total staff and were to receive 39 percent of staffing gains, and facilities that are above the range represented 53 percent of staff and were to receive 48 percent of staffing gains. These results contrast with the expectation that staffing plans would assign more staff to facilities classified as below range. The general policy that avoids placing new hires in the most difficult of the below-range, hard-to-staff facilities might account for some of this discrepancy. However, many towers of Level 4 to 86 are below the staffing range. Some of them would, presumably, be appropriate training sites for new hires, who could later transfer into higher-level facilities that are hard to staff. TRANSFERS FAA can advertise vacancies and solicit controller transfers, or individuals can make unsolicited employee requests for reassignment (ERRs, i.e., voluntary transfers). Both can help correct staffing imbalances among facilities. In addition, a small percentage of transfers are requested under “hardship” conditions, a subcategory of the ERR pool. The committee reviewed the 744 actual transfers that occurred during FY 2012, which represented about 5 percent of the workforce that year (see Table 4-6).7 From a staff management perspective, a transfer was deemed not to change facilities’ staffing levels overall if a controller moved between facilities categorized as having the same relative staffing; this was the case in 43 percent of the transfers 6 As indicated in Chapter 2, FAA air traffic facilities have several classification levels, which are based on numerous factors, including traffic volume, complexity, and sustainability of traffic. The levels range from 4 to 12, with 12 being the most complex. 7 The 744 actual transfers for FY 2012 represent a significant increase from the 563 “agreed-on” transfers included in the FY 2012 ALA–ATO staffing plan (see Table 4-4).

OCR for page 53
64 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs TABLE 4-6 Transfers Among Facilities in FY 2012 “To” Staffing Status “From” Staffing Status Below Within Above Total Below range 3 37 25 65 Within range 22 139 159 320 Above range 46 136 177 359 Total 71 312 361 744 [319 (= 3 + 139 + 177) transfers in FY 2012]. Similarly, transfers were categorized as representing good strength management when controllers moved out of facilities categorized as being above or within the staffing range into facilities categorized as below the range, or out of facilities categorized as being above the staffing range into facilities categorized as within the range; this was the case in 27 percent of the transfers [204 (= 46 + 22 + 136) transfers in FY 2012]. Finally, transfers were categorized as representing poor strength management if controllers moved out of “below” or “within” facilities into “above” facilities, or out of “below” facilities into “within” facilities; this was the case in 30 percent of the transfers [221 (= 25 + 159 + 37) transfers in FY 2012]. Only 9 percent of transfers [68 (= 22 + 46) in FY 2012] moved staff from the “within” or “above” category to the below category. The rationale behind these transfers is not apparent. Some may have been hardship cases or may have included transfers allowed as part of a progression of training to higher-level facilities. However, the overall pattern suggests the lack of a systematic process within ATO to solicit and approve voluntary transfer requests in a way that moves toward FAA’s target staffing ranges. Several difficulties and disincentives limit FAA’s ability to solicit and approve voluntary transfers. Controllers report that the costs of moving between facilities are not fully covered; this is a particularly strong disincentive when controllers are required to move their households.8 Furthermore, on arrival, the controller will be designated as a CPC-IT and required to train and qualify into the new facility. Transferring to a higher-level facility carries the risk that the controller will not qualify into the facility, with associated uncertainties concerning subsequent reassignment back to a lower-level facility. Evidence suggests that under the process by which ERRs are reviewed by individual facilities, the first facility to approve the transfer gets the controller. Thus, a facility described as having adequate staffing may get a controller who had indicated a willingness to transfer to a facility with low staffing. This process may contribute to the discrepancies between staffing targets in the hiring plan and the staffing results in the field. The composition of staff at facilities changes slowly over time. Although FAA is reported to have a substantial share of controllers who are anxious to move to higher-level facilities offering higher pay as their skills are refined, many may be reluctant to move once they have established households. To take a longer time dimension into account, the committee examined plan execution for FY 2011 through FY 2013, for which ALA provided annual comparisons of 8 Andrew LeBovidge, committee member and NATCA representative, personal communication to Safety subgroup of the committee, March 19, 2013.

OCR for page 53
Development and Implementation of Staffing Plan 65 planned versus executed staffing. The committee categorized the execution of a facility’s staffing plan as “under,” “proper,” or “over” as follows: • For a small planning value (i.e., calling for fewer than 10 staff changes at the facility), the “proper” range was set at plan ± 1. For example, a staffing plan of 5 executed at 4, 5, or 6 would be considered “proper”; it would be considered “under” if execution < 4 or “over” if execution > 6. • For a planning value ≥ 10, the “proper” range was set at plan ± 10%, analogous to how ALA defines the staffing range. For example, a staffing plan of 20 executed at 18 through 22 would be considered “proper”; it would be considered “under” if execution < 18 or “over” if execution > 22. Systemwide, the FY 2011 plan of 829 was executed at 822 and the FY 2012 plan of 961 was executed at 925. The FY 2013 plan of 1,315 was severely underexecuted at 554 because of the sequester of FAA funding. The facility counts by execution category are given in Table 4-7. Even for FY 2011 and FY 2012, which were not affected by the sequester, there was no apparent management discipline to ensure that execution was in accordance with the plan: 73 percent of the facilities “properly” executed in FY 2011, and only 52 percent did so in FY 2012. The differences between FAA’s staffing plan and execution of the plan emerged late in the committee’s deliberations, past the time when further dialogue with FAA staff might have provided explanations for the discrepancies. The committee understands that the execution TABLE 4-7 Planned Versus Executed Hires and Transfers Fiscal Year Category 2011 2012 2013 System Over 51 81 53 Proper 230 163 138 Under 34 71 124 Total 315 315 315 Centers Over 0 4 1 Proper 22 12 3 Under 1 7 19 Total 23 23 23 Terminals Over 51 77 52 Proper 208 151 135 Under 33 64 105 Total 292 292 292 NOTE: Table gives number of facilities grouped by facility type and whether the facility overexecuted, properly executed, or underexecuted its staffing plan.

OCR for page 53
66 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs process is complex and involves considerations by facility managers and higher-level staff within ATO. Furthermore, the committee does not have a full understanding of how the staffing plan is carried out and who makes final decisions on hiring and transfers. Nonetheless, because staffing levels are in flux, poor execution of the staffing plan in a particular year makes it more difficult to achieve the staffing target in a future year. The committee does not know whether the discrepancies between plans and execution are caused by improper execution or by a difference of opinion between ALA and ATO with regard to appropriate staffing levels at individual facilities. STRATEGIES TO IMPROVE FACILITY STAFFING FAA faces many challenges in steering staffing levels toward its goals. In the intensely competitive airline industry, carriers can pull out of major facilities on short notice, as in Columbus, Ohio; Memphis, Tennessee; St. Louis, Missouri; Raleigh, North Carolina; and San Jose, California, leaving behind a tower whose staff was built up over many years. Unless controllers are moved against their will, many years of attrition may be required before the facility is rightsized. Voluntary requests for transfers are placed by only a small proportion of the controller workforce, and as noted above, FAA apparently makes little effort to steer the transfer requests from facilities that are above the range to those that are below. Possible opportunities available to FAA that could facilitate the rightsizing of individual facilities are discussed in the following subsections. Career Advancement Within the Controller Workforce In theory, controllers have several paths for career advancement. Some include the transition to management and supervisory positions. Of interest here is the progression from lower-level facilities with light traffic demand and simple traffic flows to progressively higher-level facilities serving the busiest airspace with the most complex traffic. FAA’s maximum salaries for controllers appear to be structured in a way that should encourage controllers to move from lower- to higher-level facilities to achieve higher incomes. For example, a CPC who progresses from a Level 8 tower to a Level 12 facility could increase his or her maximum pay by 40 percent. However, controllers express concerns about the cost of moving households, movement into areas with higher costs of living, and the risks of transferring from a lower- to a higher-level facility and then failing to qualify.9 Despite the impediments, for those controllers eager to advance, a career progression from lower- to higher-level facilities implies that at least some lower- and midlevel facilities could not only handle their own traffic demand but also help in training controllers for higher- level facilities. Such facilities could serve as the foundation of an apprenticeship model. Staffing plans would need to allocate more developmentals and CPC-ITs from even lower-level facilities to these “training grounds” than are required by the staff-to-traffic philosophy underlying the staffing standard and staffing range described in Chapter 3. The staffing plan could involve the 9 Dean Iacopelli and Eugene Freedman, NATCA, presentation to the committee, January 2013; and Andrew LeBovidge, committee member and NATCA representative, personal communication to Safety subgroup of the committee, March 19, 2013.

OCR for page 53
Development and Implementation of Staffing Plan 67 CPCs at the training ground facilities in helping the higher-level facilities select and train suitable candidates for advancement. For a career progression to work, FAA would probably need to be more explicit with new staff about expectations concerning the need to move to advance their careers, and FAA and the National Air Traffic Controllers Association (NATCA) would need to agree on policies to induce staff movements from facilities that have ample staff to those that do not. Career progression policies might take a different approach at centers, which are large enough to afford internal staff progression opportunities. The need for an apprenticeship model was demonstrated in the cases of some high-level facilities that required new controllers because of high attrition. Under past policies, new controllers were placed directly into these facilities; the facilities were not able to advertise for and transfer in controllers who had proved capable in facilities one level down in difficulty. The resulting failure-to-qualify rates, especially at hard-to-staff facilities such as the New York TRACON, imposed a significant training burden and wasted resources, and the nonqualifiers had to be replaced in subsequent years’ staffing plans at the cost of significant delay.10 The failure rates also discouraged new controllers from indicating an interest in these facilities. More recently, ATO has avoided the placement of newly hired controllers in high-level facilities. However, ATO has no choice but to place newly hired controllers in some hard-to-staff facilities that do not attract adequate transfers, such as the New York TRACON, in the hope that some will qualify. FAA has recognized some of the concerns described above and chartered an independent review panel that provided several recommendations in 2011 on candidate selection, hiring, facility assignment, and training within the controller workforce (Barr et al. 2011). FAA, in concert with NATCA, is moving forward with many of the recommendations that are relevant to this section, particularly with regard to strengthening the initial selection and facility assignment by ATO. However, the extent to which facility assignment takes facility staffing into account relative to the staffing standard is unclear. Furthermore, the committee could find no consistent or formal policy for career advancement or for designation of training ground facilities within the staff planning processes, although an informal process appears to be in place under which certain facilities serve as training grounds for higher-level facilities. Hard-to-Staff Facilities All facilities have some level of concern with regard to selection and training, transfers, and career advancement. Certain facilities experience these concerns to such a degree that they are chronically understaffed and are considered hard to staff. In the past, FAA sent new hires into these facilities to fill gaps and allowed transfers of CPCs from low-level towers into large TRACONs and centers. The result was unacceptably high attrition, because some hard-to-staff facilities manage traffic that is among the most demanding in the nation. Recent initiatives at FAA include development and validation of operational assessments to prescreen applicants to specific facilities to maximize training success. They are still considered pilot programs and are not yet institutionalized and available to all the facilities that 10 Dean Iacopelli and Eugene Freedman, NATCA, presentation to the committee, January 2013. Rich McCormick, Gene Burdick, and David Burkholder, briefing to a subgroup of the committee at FAA headquarters, August 29, 2013.

OCR for page 53
68 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs may benefit from such a program. Additional incentives may be required to encourage more transfers into hard-to-staff facilities. Selection and Training FAA’s generation and execution of safe, efficient staffing plans are affected by its ability to select and train controllers with confidence that they will qualify into the facility. Selection and training are complicated by differences in the necessary level of skills among facilities: higher- level facilities (those handling more complex, higher-volume traffic) require special skills developed over years of training, and not all controllers may be appropriate for such facilities. New entrants to the controller workforce who have finished at the academy are not considered fully trained. Once they move to an operational facility, they are categorized as developmentals. They require additional training and must qualify on all positions in the facility before they attain CPC status. When controllers transition between facilities, they are categorized as CPC-ITs until they qualify on all positions in the new facility. CPC-ITs also require further training when they move to a higher-level facility, and they may not qualify. The training within the facility must be conducted by CPCs who are certified as trainers. Staff planning must consider the impact of placing a large number of developmentals and CPC-ITs into a facility in terms of the training burden imposed on the CPCs and the risks of high rates of attrition (Barr et al. 2011). (FAA attempts to limit the proportion of trainees at a facility to 35 percent of total personnel to avoid overburdening CPCs with training duties.) The core of controller training is conducted with the developmental or CPC-IT actively working on an operational position under the direct supervision of a CPC instructor, who is responsible for intervening in case of any problems. While such on-the-job training is necessary, committee discussions with senior FAA safety staff and National Transportation Safety Board investigators indicate that it may create a safety concern if it is not monitored properly. Thus, care must be taken to avoid overloading a facility with more developmentals and CPC-ITs than it can train safely and effectively, and the impact of on-the-job training on safety requires close monitoring. Efficient Shift Scheduling As indicated in Chapters 2 and 3, providing a scheduling tool to facilities would benefit safety (in terms of fatigue) and the generation of staffing standards and the staffing plan. The discussion in this chapter reflects the operational aspects of generating efficient schedules. Each facility generates its own schedule. Committee members have seen examples of schedules generated through the use of locally developed white boards and spreadsheets. In many cases, a default schedule template has evolved within a facility over several years and is familiar to supervisors and the controller workforce there. The operational benefits of providing a tool to facilities to assist in maximizing the efficiency of shift coverage and to clarify to the workforce at the facility how schedules are developed are described here. The benefits with regard to fatigue mitigation and staff planning will only be realized if the facility applies the tool to generate the schedules that are actually used. Such a mandate for facilities to use the tool, and perhaps some constraints on the schedules that they may select from it, must consider the operational realities of each facility. One centralized schedule cannot work for all facilities. First, traffic may be heavy or light for different facilities at different times of

OCR for page 53
Development and Implementation of Staffing Plan 69 day. Second, each facility may have local concerns such as preventing shifts from starting or ending during peak-hour traffic to avoid long commutes during recuperative breaks. Airservices Australia has completed such an effort, and positive results were reported to the committee. The organization’s scheduling capability is integrated with fatigue risk management so that schedules can be created in accordance with fatigue risk management principles and the fatigue risk of any necessary changes can be considered. With regard to providing facilities with a measure of local control and flexibility, each facility has three default schedule templates to select from. If facility management believes that none of the defaults adequately reflects its circumstances, it may propose a fourth to headquarters. As part of fatigue risk management, the scheduling tool helps identify circumstances under which, for example, offering a controller a taxi ride home after an unexpected, unusually long shift may be warranted. Thus, the scheduling tool can inform and guide the generation of efficient schedules that take into account the principles of fatigue risk management at each facility, yet each facility should be provided with the flexibility to adapt the schedules to local considerations. The schedules can have significant professional and personal impact on controllers, so they will be most effective if they are developed collaboratively with the workforce. Communication Between Operations and Staff Planning Generation of the staffing standards and staffing plan is inherently a function of central planning units in ALA and ATO at FAA headquarters. However, the facilities need a well-defined way of communicating their needs in the development of the staffing plan, particularly if unusual or unique circumstances are not reflected in the mathematical models underlying the planning process. Such communication requires clarity with regard to how the facilities’ current staffing levels were assessed and how the transition toward the staffing targets will occur. Several entities need to be involved. In addition to communication up and down the organizational chain from facility to regional service unit to centralized planning at FAA headquarters, the communication should involve the workforce at all levels. Horizontal coordination within the organization between facilities and service units would often be beneficial. Furthermore, independent review panels and working groups that are asked to examine staffing-related concerns need the ability to communicate their findings and recommendations through the organization so that improvements are broadly reflected. The committee was unable to achieve a clear understanding concerning the communication of facility concerns in the development of the staffing plan. Various processes appear to be involved, and they appear to change over time. A more consistent and transparent communication process is needed. SUMMARY AND RECOMMENDATIONS At the end of FY 2012, the majority of FAA facilities were either above (43 percent of facilities) or below (25 percent of facilities) the staffing range established in the FY 2013 controller workforce plan. A reasonable expectation is that FAA intends to move these facilities toward their staffing targets over time. After ALA uses the staffing targets to develop a hiring plan, ALA and ATO agree on a staffing plan for new hires and transfers that is intended to establish agency goals for these

OCR for page 53
70 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs facilities. The staffing plan agreed on by ALA and ATO for FY 2012 is 6.5 percent above what the modeled output suggests is appropriate. Explanations for the discrepancy included the need to account for local circumstances and facility manager judgment, but no documentation or criteria were available to the committee to justify a discrepancy of this magnitude or to indicate that the staffing standard is low. It is certainly appropriate for facilities to provide input into the staffing plan, but a transparent process is needed by which facilities can understand what the various staffing standards, targets, and ranges represent and then provide their input in an informed manner. The staff hiring determined in the ALA hiring plan and the staff planning methods themselves may be erroneous because FAA’s assessment of facility staffing does not appropriately account for non-CPC controllers. If the CEW concept described above is developed and proves effective, FAA would have better insight into a given facility’s short- and midterm staffing status. On the basis of the data the committee examined, FAA appears to execute its staffing plan as intended in the aggregate, but there are difficult-to-understand discrepancies at the level of specific facilities. For example, as indicated in Table 4-7, 73 percent of FAA’s facilities successfully implemented planned hires and transfers in FY 2011, but only 52 percent did so in FY 2012. Discrepancies between staffing plans and their execution compound the difficulty of achieving staffing goals. There was a significant discrepancy between the transfers planned for FY 2012 (563) and the number executed (744). The number of staff transfers allowed annually is roughly two-thirds the number of new hires and about 5 percent of the total workforce. Furthermore, any transfer action may require two training cycle events—one for the person transferring into a new facility and one for the person replacing the transferee. Thus, transfers can influence whether facilities move toward FAA’s staffing targets. FAA appears to lack a strategy and mechanisms to influence transfers toward this end. The committee does not fully understand how the execution of staff planning works in practice and whether it corrects staff imbalances over time. From the data available to the committee, the execution process does not appear to be doing so. This topic is worthy of further examination. There appear to be enduring differences between the staffing targets developed by ALA and the staffing in the field as executed by ATO. The differences may be caused by poor execution of the annual staffing plans or by a difference of opinion within FAA with regard to the appropriate level of staffing at individual facilities. Recommendation 4-1. FAA should examine the merits of using a more appropriate analysis of facility demographics (e.g., the CEW concept) in place of the CPC + CPC-IT metric when it assesses facility staffing status and develops its annual staffing plans. Recommendation 4-2. FAA should develop explicit criteria with regard to when and why the staffing plan for a given year can exceed the hiring plan based on the staffing standard. Recommendation 4-3. FAA should ensure that the field understands how facilities’ staffing levels (current and target) are assessed and offer a transparent process by which facilities can provide input on facility staffing levels established in the annual staffing plan.

OCR for page 53
Development and Implementation of Staffing Plan 71 Recommendation 4-4. FAA should make more effective use of voluntary transfers to rebalance the workforce among facilities considered to have high staffing levels and those considered to have low levels, particularly where it can leverage controllers’ desire to transfer on the basis of hardship, career advancement, or personal circumstances. Such a strategic process would need to include the following: • Suitable incentives for transfers, developed and agreed on by FAA and NATCA, that would help rectify staffing imbalances, including establishment of a policy for resolving situations in which controllers who are willing to risk transferring to a higher-level facility fail to qualify into that facility; and • Systemwide processes and management of transfer requests to consider their impact on facility staffing, so that facilities with low staffing levels do not miss voluntary transfers claimed first by facilities with adequate or high staffing levels. Recommendation 4-5. FAA should establish and advertise a clear path for the career advancement of controllers. They should be mentored and should understand the expectations on which their advancement will depend, and those expectations should be understood in staff planning and at the training ground facilities. Recommendation 4-6. FAA should adopt a formal apprenticeship model that reflects the potential of lower-level facilities to serve as training grounds and sources for transfers into higher-level facilities. An informal system appears to be at work across some facilities. A more formal and well-communicated model may work more effectively. Lower-level facilities would need to be staffed in a way that reflects their dual role of serving traffic at the facility and bearing a heavy training burden. FAA and NATCA will need to work together to develop incentives to keep controllers moving upward through the system at an appropriate rate. To function effectively, such an apprenticeship model must be transparent to the workforce. Clear designation of the facilities is a first step. Controllers should be mentored with regard to furthering their career development. Standard policies or documentation should be available to controllers indicating how long they might be placed in lower- and midlevel facilities and when they would be expected to transfer to higher-level facilities. Controllers would be more motivated to participate in career development if selection processes were in place to reduce the risk that they might not qualify into a new, higher-level facility and if clear policies were in place for handling their reassignment should they not qualify. Recommendation 4-7. FAA should work with NATCA in developing and implementing special measures to address the concerns of hard-to-staff facilities. Such measures might include • Incentives to transfer to and reside near the facility; • Selection processes and policies that would prescreen controllers willing to transfer to facilities and identify controllers likely to qualify (and not qualify); • Policies for reassignment of trainees who fail to qualify, to remove disincentives for potential applicants to the facility; and • Methods for training controllers at facilities one level down in difficulty to nearly the level required at the hard-to-staff facility, thus minimizing the training requirement once they are on site.

OCR for page 53
72 Federal Aviation Administration’s Approach for Determining Future Air Traffic Controller Staffing Needs Recommendation 4-8. In its staff planning, FAA should consider the impact of placing a large number of developmental controllers and CPC-ITs into a facility in terms of the training burden on the facility. FAA should monitor for any safety concerns arising with intensive on-the-job training within facilities. Workforce, facility, and service units should be able to communicate concerns as feedback into the staff planning process. Recommendation 4-9. FAA and NATCA should collaborate in implementing a scheduling tool and procedures for its use at each facility. They should establish standard schedule templates suitable for each facility’s traffic patterns. More than one standard schedule should be available to each facility, and a “request for further consideration of unique circumstances” process should be available to facilities who believe that local circumstances require adjustment to their schedule templates. The scheduling tool and associated procedures should allow each facility to design, revise, and publish schedules that take best practices in efficient shift scheduling and fatigue mitigation into consideration. The scheduling templates used at the facility should be transparent and available to others, including those performing the staffing process. This may be fostered by the scheduling tool having the capability of archiving the scheduling templates in a manner that allows different organizations to access them. REFERENCES Abbreviation FAA Federal Aviation Administration Barr, M., T. Brady, G. Koleszar, M. New, and J. Pounds. 2011. FAA Independent Review Panel on the Selection, Assignment and Training of Air Traffic Control Specialists. Final report. Sept. 22. 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.