5

Implementation and Sustainability of the Staffing Model

The success of any staffing model is dependent on both the validity of the model developed and the manner in which it is implemented. The purpose of this chapter is to present and discuss both the requirements for successfully implementing a staffing model and the factors that must be considered to sustain the use of the model over time.

TIMELINE

Airway Transportation Systems Specialists (ATSS) are responsible for maintaining many types of equipment in multiple types of locations across a wide geographic area, and as discussed in previous chapters, many factors affect the staffing levels necessary to maintain the National Airspace System (NAS) effectively and safely. Consequently, any model that will accurately estimate necessary ATSS staffing levels will be complex and likely require a considerable amount of time to develop. Similarly, a complex staffing model is likely to need a great deal of preparation for implementation, ranging from testing the model to training Federal Aviation Administration (FAA) personnel to use the model, and it is likely to take a considerable amount of time to implement.

The actual time required to develop such a model should be proposed by the developer. Experts on the committee who have developed complex models estimate the development time to be between 12 and 24 months for a deterministic model and longer for a robust simulation tool like the Air Force’s Logistics Composite Model. The implementation time could take 6-12 months in conjunction with the FAA’s planning, programming, and budget cycle requirements. If there are significant changes to staffing at locations, these may need to be managed over a longer time through attrition, hiring lead times, etc. The FAA’s 2012 contract with the Professional Aviation Safety Specialists (PASS) labor union, maintains the 6,100 ATSS staffing level as a staffing floor for 16 months (Professional Aviation Safety Specialists, 2012). If the developer’s schedule estimates for model development and implementation exceed the 16-month limit, the FAA will need to determine how to manage staffing levels until the new staffing model is implemented.



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 79
5 Implementation and Sustainability of the Staffing Model The success of any staffing model is dependent on both the validity of the model developed and the manner in which it is implemented. The purpose of this chapter is to present and discuss both the requirements for successfully implementing a staffing model and the factors that must be considered to sustain the use of the model over time. TIMELINE Airway Transportation Systems Specialists (ATSS) are responsible for maintaining many types of equipment in multiple types of locations across a wide geographic area, and as discussed in previous chapters, many factors affect the staffing levels necessary to maintain the National Airspace System (NAS) effectively and safely. Consequently, any model that will accurately estimate necessary ATSS staffing levels will be complex and likely require a considerable amount of time to develop. Similarly, a complex staffing model is likely to need a great deal of preparation for implementation, ranging from testing the model to training Federal Aviation Administration (FAA) personnel to use the model, and it is likely to take a considerable amount of time to implement. The actual time required to develop such a model should be proposed by the developer. Experts on the committee who have developed complex models estimate the development time to be between 12 and 24 months for a deterministic model and longer for a robust simulation tool like the Air Force’s Logistics Composite Model. The implementation time could take 6-12 months in conjunction with the FAA’s planning, programming, and budget cycle requirements. If there are significant changes to staff- ing at locations, these may need to be managed over a longer time through attrition, hiring lead times, etc. The FAA’s 2012 contract with the Professional Aviation Safety Specialists (PASS) labor union, maintains the 6,100 ATSS staffing level as a staffing floor for 16 months (Professional Aviation Safety Specialists, 2012). If the developer’s schedule estimates for model development and implementation exceed the 16-month limit, the FAA will need to determine how to manage staffing levels until the new staffing model is implemented. 79

OCR for page 79
80 ASSESSMENT OF STAFFING NEEDS OF SYSTEMS SPECIALISTS IN AVIATION 1) Staffing Model Development/ Amendment 6) Monitoring of 2) Staffing Model Staffing Model Pretest Results 3) Preparation for 5) Staffing Model Staffing Model Validation Implementation 4) Staffing Model Implementation FIGURE 5-1 Phases of staffing model development and implementation. 5-1.eps Figure 5-1 provides a high-level overview of the major steps from model development through imple- mentation. Each of these steps comprises a number of activities that should be undertaken to increase the likelihood of a successful model launch. Each activity requires time and resources. Ideally, the FAA and its staffing model experts would add to these major steps as needed and elaborate on each, specifying what needs to be done by whom and creating a schedule to be followed for each step. Step 1 in Figure 5-1 includes all the activities associated with the actual development of the model that have been detailed in the preceding chapters. Specifically, Figure 3-1 in Chapter 3 details the five phases that should be incorporated into model development. Although the workforce planning process described in Chapter 3 and the implementation process discussed here in Chapter 5 overlap considerably, this discussion is meant to highlight the activities related to successful implementation. Although much of the work falls on the shoulders of the experts designing the model, the input into the model will likely be the responsibility of FAA employees. The accuracy and the value of the results of the model are limited by the quality of that input. Inaccurate, biased, or incomplete data can skew the recommended staffing levels. Data from such time recording systems as the Labor Distribution Record- ing system are noted for their limitations and the unreliability of their data. Similarly, information about the time required for maintenance may be highly variable due to the age of the equipment, the amount it is used, and the environment in which it is used. The FAA staff will need to participate in activities aimed at developing processes that produce accurate information about the tasks ATSS personnel per- form and the factors that affect the time required to perform them. Ideally, in the long term, the FAA

OCR for page 79
IMPLEMENTATION AND SUSTAINABILITY OF THE STAFFING MODEL 81 would embrace information technology enterprise architecture strategies. A comprehensive enterprise architecture would encompass a visionary review and where appropriate, integration of the entire set of information technology systems that ATSS workers and their leadership find necessary for data input. Such an architecture would boost future modeling and total resource control, including control of staffing. To begin, the FAA may desire to modernize human resource, time tracking, and maintenance systems, and at the same time integrate them to achieve synergies between these systems, because they are used by the ATSS workforce, as well as to oversee and manage it.1 The introduction of a staffing model offers an opportunity to establish productive relationships among the people critical to developing, implementing, and using it, as well as an opportunity to create systems that interact and support each other, limiting the need for redundant inputs of the same infor- mation. The modeling experts, data experts, and ATSS functional experts could seize this opportunity to form an enduring partnership and ensure communications intended to constantly improve the model, its inputs, and its utility and usability. In addition, the need for an accurate time reporting data base that feeds the staffing model creates opportunities to improve and integrate human resource systems. Moreover, the more that FAA employees are involved in the development process—as opposed to simply arriving one day to learn that a new staffing model is in place—the more likely they will be to provide the inputs needed to make the model work well and the more they will be to have an “owner- ship interest” in the model and be willing to help with the inevitable tweaking process as the model is implemented. For the employees to participate willingly, they must understand that the purpose of the model is to improve their ability to perform their jobs more efficiently and effectively, rather than, for example, to see how much the workforce can be reduced or increased. There are numerous Safety and Risk Management protocols available today to help quantify potential risks both objectively and subjectively and then prioritize or target those risks that appear unacceptable or too great not to take proactive measures to manage them better. Current FAA Safety and Risk Manage­ ment frameworks and safety metrics may provide a starting point to assist in providing guidelines to a specific modeling effort. Still, it is highly likely that a team of functional experts—to include those with significant expertise in assessing and improving safety and decreasing risk—would be useful if called on to help clarify risks inherent in processes and staffing associated with a particular study effort. Step 2 involves a pretest of the model prior to full implementation. The purpose of this type of pretest is to identify problems with the model and its results and rectify them prior to actual model use. The basis for this evaluation is the set of quality factors described in Step 5. Because the evaluation in Step 2 is preliminary and precedes the roll-out to users, several aspects cannot be assessed. For example, usability cannot be studied directly as the ultimate users will not actually use the model prior to the Step 2 evaluation; however, pilot tests with samples of users should be encouraged. Those who create the model need to generate a “trial model application,” compare the results to current (or past) staffing levels, and collect data about the impact of the new staffing levels (whether they are higher or lower or remain the same) on the FAA’s ability to maintain the NAS. For this test of the model, in addition to collecting hard data for statistical comparisons, such as number and length of outages, overtime hours, and backlog, the modelers need to prepare to collect qualitative data such as leaders’ reactions to the differences between actual and proposed staffing levels. Chapter 3 provides five criteria by which a staffing model should be evaluated: transparency, scal- ability, usability, relevance, and validity. The new ATSS staffing model should be compared to these standards and adapted as needed prior to implementation in Step 4. As review and improvement of the 1FAA NAS Enterprise Architecture. Presentation by Jesse Wjintjes, NAS chief architect, to the April 28, 2010, Meeting of Agency Chief Architects. Available: http://www.jpdo.gov/library/PartnerAgency/Jesse_Wijntjes_Briefing.pdf [May 2013].

OCR for page 79
82 ASSESSMENT OF STAFFING NEEDS OF SYSTEMS SPECIALISTS IN AVIATION staffing model is a continuous process, these five criteria should be kept in mind once the model is implemented and is being validated and amended. Step 3 focuses on the activities necessary to prepare the FAA to transition from the current state of staffing to a new process for establishing staffing levels via a carefully constructed staffing model. This committee was specifically charged with considering transition in its statement of task (see Box 1-1). Because the current staffing level for ATSS personnel of 6,100 is dictated by the contract to which the FAA and PASS agreed (Professional Aviation Safety Specialists, 2012), the transition phase will not focus on changing the procedures for estimating staffing requirements and allocation of those personnel. Instead, the transition should focus on installing the new modeling system, procuring equipment (e.g., computer equipment to run the model and distribute reports), personnel, and other resources needed to execute the model, ensure reliable data inputs, and educate users. In addition, the FAA will need to develop a detailed plan, based upon input from the employees, to implement the new staffing model. For example, the FAA must decide whether the staffing model should be rolled out across the country at one time or implemented region by region, determining the pros and cons to each approach and remediating as many of the disadvantages of the option chosen as possible. In addition, the FAA must develop the timeline for the transition, checking to see what other events might affect the implementation. This implementation plan also needs to contend with the implications of decreased numbers of ATSS personnel needed in some locations and increased numbers of ATSS personnel needed in other areas. Among such implications are preparing the organization for employee movement (e.g., movement of ATSS personnel from areas where their expertise is surplus to other areas where it is needed), employee redundancy, and new hiring demands. If the employee movements are significant, they can be phased in over many fiscal quarters or several years. To the extent that person- nel will be moved in ways that require bargaining with the labor union representing ATSS personnel, the FAA should prepare for and fund contract negotiations with PASS. The current hiring process for ATSS personnel and the training and certification processes are lengthy. When ATSS personnel are needed in one location, the FAA must factor the hiring-training cycle into its plans. Although the task may be daunting, the FAA might also consider improvements to the new hire processes as it develops and implements a new staffing model. In addition, the transition plan should include training of both the staff who execute the model and all the managers and ATSS technicians who are affected by it. A new staffing model should be developed by model experts in partnership with functional experts and staff. It will be an essential requirement of implementation to train those who will use it so that all levels can use the reporting mechanisms of the model. Because of the number of users and their varied needs, training FAA personnel in how to use the system and interpret the output that is relevant to their work will not be a trivial undertaking, and the personnel necessary to develop and provide the training, as well as support ongoing usage of the modeling system, must be funded. The model is likely to be better accepted if affected personnel (i.e., ATSS personnel and their manag- ers) are informed of how the model works. This is part of transparency, one of the quality criteria from the 2006 report, Staffing Standards for Aviation Safety Inspectors (National Research Council, 2006). An understanding of the statistical intricacies of the model is probably not necessary for most employees; however, an overview of the input and the considerations that the model takes into account will help managers and ATSS personnel see how numbers are derived, appreciate the rationale behind increases and decreases in staffing levels, and understand their role in providing quality inputs into FAA data systems. This will also help FAA employees to be understanding contributors to periodic model refine- ments and updates—for the good of the whole team and the system they support. In addition, PASS must

OCR for page 79
IMPLEMENTATION AND SUSTAINABILITY OF THE STAFFING MODEL 83 understand the basics of the model and support its development. Without PASS’s active participation, it is unlikely that a sufficient model can be built or implemented. Step 4 encompasses the actual implementation and rollout of the model according to the plans laid out in Step 3, all of which should logically be synchronized with key FAA human resources, training, and budget processes to the extent possible. It is a tool to help all of these systems to be more accurate and effective by defining and defending the full-time equivalent requirements needed to serve the NAS. Step 5 is the evaluation of the quality of the staffing model against the five standards discussed in Chapter 3: relevance, scalability, transparency, usability, and validity. Once the model is implemented, the modeling team must determine how well the model and its implementation are doing on the follow- ing markers of quality: • Relevance: Does the model capture the important drivers of workload and provide the appropriate level of detail in the output? • Scalability: Can the results of the model be accurately aggregated at higher levels to predict staff- ing needs by facility, geographic region, or discipline? • Transparency: Is the model understood by the users and those affected by it? • Usability: Can the model be implemented and improved to make the intended staffing predictions by those who were intended to use it? • Validity: Does the staffing level predicted match the actual need and allow the ATSS workforce to maintain the NAS safely and efficiently? All of the quality standards are important, but validity is perhaps the most critical. Unless the results of the staffing plan allow the FAA to maintain the NAS, nothing else is relevant. Also important is the sustainability of the staffing system. If tasks associated with using the model are so laborious that they cannot be accomplished in a reasonable period of time, the modeling system is unlikely to be used and maintained regardless of its validity. To a large extent Step 5 and Step 2 are similar. Both are designed to evaluate the model’s quality. The evaluation in Step 5 is usually more extensive because it occurs after the model has been used. Step 6 includes the activities that are necessary to sustain the model over a period of time: monitor the staffing model, evaluate its adequacy, and make adjustments as needed. The staffing model that is produced as a result of this study will be more effective if it is periodically reviewed and updated to reflect the changing environment of ATSS work. The FAA should try to identify as many sources of information about the effectiveness of the staff- ing model as possible. While much information will come from the professionals who use the model to estimate and allocate ATSS personnel, the FAA should consider identifying best practices from other large organizations that use staffing models, as well as creating employee suggestion programs and lessons-learned programs that encourage constructive criticism and helpful suggestions for remediation. Both quantitative and qualitative data can inform the FAA of the strengths and weaknesses of the staffing model. Hard data such as outages, overtime, and inspection and preventive maintenance backlogs relate to the ultimate criterion: maintaining staffing at levels necessary to keep the NAS safe and efficient. Qualitative data such as supervisors’ perspectives on revised staffing levels, including the effect on employee fatigue and morale, can also help to evaluate the effectiveness of the model as well as detect unintended consequences from applying it. Note that the implementation cycle begins again after Step 6. As results from the FAA’s ongoing evaluation against the five quality standards indicate where amendments are needed to the staffing model, modifications should be made and the cycle of implementation should begin anew. An effective

OCR for page 79
84 ASSESSMENT OF STAFFING NEEDS OF SYSTEMS SPECIALISTS IN AVIATION and useful modeling process takes into account the lessons that have accrued from each round of the implementation process and incorporates them into the next iteration of the staffing model, resulting in a spiral development process. During this cycle, one change related to the modeling parameters may be the result of several changes related to the work performed by ATSS personnel. For example, by using the model through one staffing cycle, the modelers and the users of the model may learn that certain forms of data are not reliable enough to produce sufficiently accurate estimates of staffing needs. The inaccuracy of the data may be the result of aging equipment that demands varying amounts of time for maintenance and repair. At the same time, the quality of the data may be affected by the time record- ing system. Thus, changes to the model may be based on the retirement of legacy equipment as well as revisions to the time recording systems currently in use. The timelines for each step of the process are different across organizations and depend on multiple factors such as the variability of the job across locations and functions, complexity of the model designed, the number of people who provide informa- tion or must be informed, the resources available, the capabilities of the contractor who develops the model, and other factors. Nevertheless, an important first step in the design of a staffing model project will be to create a timeline that specifies the amount of time to be devoted to each stage of the project. Recommendation 5-1: The FAA should prepare a timeline that details all the activities associated with model development and implementation that must be completed and should ensure that the resources necessary to accomplish each are available. Recommendation 5-2: Once implemented, the FAA should continue to monitor the effectiveness of the staffing model by collecting data from multiple sources and should make adjustments as needed to enhance the accuracy of the model. In addition, the model should be adapted as changes to the major components of the model are made, such as changes in the tasks performed, equipment used, and train- ing processes. In addition to testing the model and engaging in continuous improvement activities, there are several critical elements related to implementation: the FAA staff who will administer the staffing model; equip- ment and other resources; and funding for all aspects of the model including consultants, equipment, FAA staff, training, and other costs. These critical elements are briefly discussed below. FAA STAFF FAA employees will be involved in the development and implementation of a staffing model in a variety of roles. Even when experts employed by a vendor develop a complex staffing model, the FAA needs to provide staff to guide the project and facilitate access to appropriate data and information to support the development work. Unless the FAA staff provides input, or assigns functional experts to periodically work with or as part of the modeling effort, the model that is created is not likely to take into account accurately the important factors that affect staffing levels and to reflect actual staffing needs. Similarly, FAA employees are likely to be involved in the actual execution of the model to establish staffing levels. Again, even if significant work associated with the implementation is contracted to a vendor, the FAA staff still needs to supply the input to the model, to ensure its accuracy and its thor- oughness. The FAA will need to plan for these positions and take the steps necessary to ensure they are filled with personnel with the appropriate knowledge, skills, and abilities.

OCR for page 79
IMPLEMENTATION AND SUSTAINABILITY OF THE STAFFING MODEL 85 Finally, the FAA staff provides the necessary continuity and historical knowledge for keeping the model current and transparent. Without knowledge of how the model was built and has changed over time, the use of any model becomes a mere rote procedure, performed ritually rather than used for its original and important purpose. In addition to the creation and actual execution of the model, there are other tasks that should be performed during the implementation phase. For example, training of the managers and ATSS personnel who are affected by the new staffing model will require training personnel who are able to explain both the model and the implications of its use, not just the mechanics of input and output. EQUIPMENT AND OTHER RESOURCES In addition to personnel, other resources related to the implementation of the staffing model include the computer systems necessary to run the models. The 2012 Grant Thornton report recommends a Web- based system that allows for broad access and facilitates system updates (Grant Thornton, 2012:28). For example, the Web-based system should allow multiple, simultaneous users while maintaining control of the processes centrally. Modules that can be turned on or off, removed, and replaced will contribute to the flexibility of a complex modeling system. Ideally, the Web-based system is updated continuously via links to data feeds from other FAA systems. For example, a data base that maintains records of main- tenance events, repairs, time spent, travel time, etc., could automatically feed and update the staffing model and revise predictions. Development of this system would need to be funded and implemented as part of the model development, if this choice is approved. FUNDING The investment in a staffing model is likely to be substantial. The FAA must be prepared to pay for not only the consultants who develop it but also for the information technology systems on which the resulting model is run, the FAA employees who interface with the consultants, and the training of FAA employees who use the model. Additionally, the FAA must be prepared either to fund the number of employees generated by the model or, if not funded, to clearly define and potentially live with the risks taken on. The agency should also be prepared to explain how these risks would be prioritized and mitigated. The funding requirement could be particularly challenging if the number of ATSS personnel determined from the staffing model exceeds the current staffing level of 6,100. The funding must cover not only the salaries of additional employees but also the processes by which these new employees are acquired and trained. There are also costs associated with downsizing and relocating ATSS personnel where they are needed, and these costs should be budgeted if that need arises. The development and implementation of a staffing model will not occur in a vacuum. There are undoubtedly many other process improvements related to the ATSS job that could be pursued and that will have an effect on the ATSS personnel workload and the staffing model. The importance of human- systems integration (HSI) was discussed in Chapters 1 and 3. As the FAA improves its operations in the manpower domain of HSI, it will need to seek continued, deliberate improvement and synergy across the other eight HSI domains. As the FAA implements a new staffing model, it has an opportunity to evaluate its entire talent management system and optimize each component. For example, training and certification procedures should be considered at the same time that a new staffing model is being devel- oped and implemented. Reducing the time to certify technicians, either through more accessible training or by the ability to certify experienced hires without travel to Oklahoma City, will affect the model’s effectiveness in maintaining the efficiency and safety of the NAS.

OCR for page 79
86 ASSESSMENT OF STAFFING NEEDS OF SYSTEMS SPECIALISTS IN AVIATION Similarly, revising the process by which the appropriate staffing number is determined will empha- size the importance of accurate input data and encourage a revision of the processes by which ATSS personnel’s time is tracked, equipment inventoried, spare parts for the NAS maintained, etc. Process improvements in how maintenance work is performed are expected to be ongoing, with continuous improvements. All of these ancillary activities will also require funding in addition to what has been allocated for the modeling effort. This report has noted repeatedly the importance of informing ATSS and PASS about the new model and explaining how it was developed, what it takes into account, how it will be implemented, etc. Funds must be available for the staff that creates written informational brochures, designs talking points, deliv- ers this information in an understandable way, and handles the questions that will arise. Because of the importance of communications to the success of a new staffing model, professionals should be engaged to maximize the clarity and persuasiveness of the material. CONCLUSION The FAA is on the right track by aggressively seeking a new and appropriate manpower model. The FAA’s contract with Grant Thornton is commendable (Grant Thornton, 2012), but it can be taken to a better level through deliberate analysis and inclusion of stochastic and performance dimensions in modeling a required workforce, even as the ATSS workforce takes on the challenge of maintaining the NAS in the coming decades. The 2-year process that FAA has already undertaken to assess its current models and data systems and to plan ways to improve its data systems should facilitate model develop- ment and implementation. Ideally, the FAA will institute a rigorous implementation process for the new staffing model and related processes and will embark on the journey of continuous improvement. In the future, the staffing system will be regularly reviewed and improved at the same time as other process improvements that will increase both the efficiency and the safety of the NAS. Implementation of a new staffing model includes a number of tasks, including activities such as acquiring computer equipment to host the model, educating users, and evaluating the model against the criteria for quality multiple times. A robust staff- ing model is likely to improve the process for allocating staff and contribute to the maintenance of the NAS; however, the implementation and use of that staffing model will require considerable attention and resources from the FAA on an ongoing basis.