Workforce Analytics for Government Agencies Organizational Planning for Sustainability and Effectiveness

WHITE PAPER

SAS White Paper

Table of Contents Executive Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Government Agencies Face a Workforce Tsunami. Are You Ready? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Workforce Analytics: Empowering Government Agencies to Optimize Staffing and Costs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Key Components of Workforce Analytics Solutions . . . . . . . . . . . . . . 5 A Solid Data Repository: Laying the Foundation for Effective Workforce Analytics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 SAS® Solutions for data integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 SAS® Solutions for data cleansing and quality . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Workforce Analytics: Turning Personnel Data into Information and Insight. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Data mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Using SAS® Solutions for data mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Predictive modeling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Using SAS® Solutions for predictive modeling . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Forecasting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Using SAS® Solutions for forecasting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Simulation and optimization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Using SAS® Solutions for simulation and optimization. . . . . . . . . . . . . . . . . . . . 12

SAS: Keeping You on the Cutting Edge of Decision Support Solutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Contributor: Jon Lemon, Solutions Specialist Federal Workforce Analytics, SAS .

Workforce Analytics for Government Agencies

Executive Summary For the past several years, the federal government workforce has been growing at a steady pace. Moreover, thousands of baby boomers have staffed key positions across all agencies; their years of experience and accumulated expertise have afforded the federal government a level of stability that has kept training and recruiting costs down, helped to ensure a consistent service level for the people they serve, and more. But in the near future, everything will change because of two key trends: • Many government agencies are faced with an aging, skilled workforce that will soon retire, taking its skills and knowledge with it. • Significant budget cuts that are expected to be enacted in the near future, which will necessitate reductions in force. Members of the both the House and Senate have recently introduced bills that call for a reduction in the federal workforce by 10 to 15 percent from 2015 to 2020. Unless agency executives plan and take action in a very deliberate, strategic way, the result will be organizational upheaval and resource gaps that pose a grave threat to agency missions. So what are executives to do? How can they look into the future and make informed plans and decisions that will ultimately ensure their agency continues to have the right staff with the right skill sets and experience to fulfill their organization’s mission? By analyzing data, managers can make better decisions that help them prepare for the looming workforce crisis. Analytics can help executives answer questions such as: • Where are we? • What’s worked in the past? • How did we get to this point? • How can we learn from the past to help shape a better future? • What skill sets and levels of experience do we need to maintain at different budget levels so we can continue our organizational mission? Past data is a wonderful source of information – and in the world of human resources, there is no lack of data. However, in most cases, data is spread over multiple systems, too voluminous to work with, and is very complex to the point where the data is simply facts and records and not fused together to make information. As explored in this paper, that’s why executives need a robust workforce analytical solution that can help them quickly aggregate and analyze massive volumes of data that would otherwise quickly overwhelm and mire down analysts and senior executives with information technology details. At the same time, workforce analytic solutions can empower them to make the best possible decisions with confidence – even when faced with highly complex scenarios – and preserve the future of the federal workforce.

1

SAS White Paper

Government Agencies Face a Workforce Tsunami. Are You Ready? Over the past several years, most government agencies have been growing and enjoyed relative stability from an organizational perspective, as key positions have been staffed by experienced “baby boomers” who are choosing to finish their careers in government (see Figure 1). And it would be easy for this historical stability to lull executives into thinking that the status quo will continue through a familiar pattern of attrition and rehiring.

Figure 1: The number of US federal government employees has remained historically stable. Source: FedScope, Office of Personnel Management (fedscope.opm.gov/ employment.asp)

But over the next few years, two macro-trends will wreak havoc with the status quo. First, the current and future economic environment is forcing federal organizations to increasingly be nimble and agile so they can adjust to constant change. “Sequestration” and “austerity” are terms used to describe predictions of near-term massive budget cuts and federal workforce adjustments. For example, bills already introduced in both the House and Senate call for a reduction in the federal workforce by 10 percent over 10 years through attrition. So it’s critical that agency executives have a way to aggregate data quickly, use analytics to turn this data into information, and then use this information to perform what-if analysis for future decision making. As organizations improve how they collect the data, it gets easier for them to turn that data into information and run more accurate what-if analyses, which in turn leads to better, more data-driven decisions that are factually- and statistically-based. Equally important, they can provide a detailed justification for final decisions if they are ever questioned. The second major factor is the bursting of the “baby boomer” bubble. For years, the federal government has known that 35 percent of the federal workforce is eligible to retire by 2016. This means federal agencies will need succession and recruiting plans for an estimated 850,000 employees – a huge task.

2

Workforce Analytics for Government Agencies

Senior executive service and career leadership eligible to retire from 2012 to 2016

42% 58%

Nonleadership eligible to retire from 2012 to 2016

30% 70% Eligible to retire from 2012 to 2016 (30%) Nonretirement eligible from 2012 to 2016

Figure 2: Baby boomers in the federal workforce. Sources: • Total 2010 US federal government employees (nonuniform military)1 • Senior executive service and career leadership eligible to retire between 2012 and 20162 • Eligible to retire from 2012 to 2016 (30%)3

At this point, the mass exodus is not a question of “if” it will happen, but rather “when.” In most cases, those individuals who are eligible to retire are occupying seniorlevel positions and have years of institutional and corporate knowledge about their organization, making their loss particularly devastating to agencies. In fact, among the 58 percent representing senior executive service and career leadership, 30 percent are eligible to retire between now and 2016 (see Figure 2). If not properly planned for, the loss of knowledge and experienced leadership these individuals represent could seriously hinder the ability of agencies to carry out their missions. The consequences of this phenomenon will be compounded further within the federal government because up-and-coming senior-level managers being groomed as replacements are increasingly being tempted by the private sector to leave for more lucrative opportunities.

1

“Historical Federal Workforce Tables – Total Government Employment Since 1962.” US Office of Personnel and Management. Retrieved from: opm.gov/feddata/HistoricalTables/TotalGovernmentSince1962.asp.

2

“GAO, OPM: IT skills gap remains a critical concern.” FCW, 1105 Government Information Group, Sept. 20, 2012. Retrieved from: fcw.com/articles/2012/09/20/berry-dodaro-addressing-critical-skills-gaps. aspx?s=fcwdaily_210912.

3

Ibid.

3

SAS White Paper

With more than 70 percent of today’s workforce comprised of knowledge workers, the ability to effectively manage the organization’s investment in “human capital” can spell the difference between success and failure. To manage effectively given these macro trends, agency executives must be able to forecast trends and make swift, confident and accurate decisions regarding highly complex, dynamic staffing issues. For example, if they are asked to cut 10 percent of their workforce, they need to make these cuts with great precision, preserving a sufficient number of people with the right skill sets and management experience to continue operating their agency effectively. The tendency may be to do an across-the-board 10 percent cut without regard to skills, mission or budget effectiveness. Or they may apply a “LIFO” (last in – first out) reduction, which mandates that cuts come from the last people hired until the targeted reduction in force is met. This methodology does not take into account those employees who are the best performers or hardest workers. Nor does it allow for succession planning or any meaningful analysis of the workforce relative to an agency’s mission. It’s just a generic, mathematical calculation applied to a very complex problem – without concern for potentially deep, long-term consequences.

Workforce Analytics: Empowering Government Agencies to Optimize Staffing and Costs To combat this potential and sudden loss of talent, organizations need a way to analyze their needs and workforce composition in flexible, dynamic ways, and use the insights gained to make confident, informed decisions that preserve the right resources at the right time. Moreover, they need an analytics solution that empowers executives to slice, dice and analyze massive volumes of data quickly and efficiently. At the same time, agencies need to get their analysis – which should ultimately drive their decisions – right the first time. Reorganizations cost money; the estimated cost of turnover is approximately 150 percent of the salary of those employees who voluntarily leave. The cost rises to 200 to 250 percent for managerial positions – and these figures do not even account for the intellectual capital that is lost.4 Equally important, reorganizations put key people at risk of leaving, resulting in even deeper staffing issues.

4

William G. Bliss, The Advisor, “Cost of Employee Turnover.” Retrieved from: isquare.com/turnover.cfm.

4

Workforce Analytics for Government Agencies

What’s needed is a robust workforce analytical solution that can help government agencies optimize staffing and costs – even as things change. These solutions can: • Gather and integrate as much data from as many date sources as possible to inform and triangulate changing workforce problems. • Use data mining capabilities on past and present employee data to discover: »» The behaviors of the workforce. »» What motivates the workforce. »» The risk of employees taking certain actions (such as leaving or staying as a civil servant, seeking detail opportunities and/or promotions, and more). »» Hidden trends and patterns in what is causing the behaviors and actions of the workforce. • Employ multiple, advanced forecasting techniques to analyze data and generate the best forecasting model possible for a particular slice of data. • Enable people to run what-if scenarios to “test” particular policy changes, understand financial impacts of proposed strategies, and/or weigh the cost and benefits of different actions. • Provide a mechanism to optimize the allocations of human capital based on constraints, such as financial, political and union constraints.

Key Components of Workforce Analytics Solutions There are two key components of any workforce analytics solution: • A data repository that unifies all relevant, yet disparate, data sources maintains intelligent linkages between them and uses them to help you quickly begin to apply analytics to get the answers you’re looking for when making labor force decisions. • A workforce analytics solution that turns personnel data into information for human resource decision making. Let’s take a closer look at the types of technologies and methods that can be employed to help the federal government prepare for the upcoming workforce predicament. As we’ll see, SAS offers workforce planning solutions that help executives manage their workforce today, while anticipating and planning for the changes and demands of tomorrow.

What sets SAS® workforce management solutions apart? • Repeatable processes: Reduce manual, error-prone activities with automated, repeatable processes for data gathering and analysis. • Flexible forecast generation: Run forecasts automatically or on an ad hoc basis whenever there’s a change in predicted demand for a program or available talent. • Automated forecasting models: Quickly generate large quantities of high-quality forecasts so executives can plan more effectively for the future. • Advanced predictive capabilities: Determine what will happen, the best that can happen and the next best action to take using advanced, predictive analytics. • Superior data management: Consolidate disparate data sources (e.g., performance evaluations, training and compensation) for a holistic view of talent.

5

SAS White Paper

A Solid Data Repository: Laying the Foundation for Effective Workforce Analytics As with any analytics solution, bad data results in inaccurate, incomplete insight – hardly useful for workforce analysts and senior decision makers. In most government organizations, data exists in many different databases and repositories. Human resources (HR) systems, time and attendance systems, general ledger financial systems, employee performance systems, and more all contain critical pieces of data that, when put together, provide a robust data set that people can analyze and gain the critical insights needed to manage people and entire organizations more effectively. That’s why workforce analytical solutions need to run on a solid data foundation powered by world-class data integration technologies that ensure data integration and data quality. Ideally, these technologies need a way to maintain intelligent linkages between disparate data sources – linkages that can be used to quickly apply analytics and provide people with the answers they need when making labor force decisions.

SAS® Solutions for data integration With SAS data quality and integration tools, you can merge any set of HR data sources and have a complete data set to analyze and connect the dots. After melding data such as employee bio records, time and attendance facts, general ledger numbers, and employee performance figures, agencies can apply analytics to integrate and then correlate the data points to reveal the larger picture needed to make informed workforce plans and decisions. For example, time and attendance and work schedule data can be merged with performance data to investigate if compressed work schedules and/or telework programs have any effect – positive or negative – on employee performance. As discussed further in the next section of the paper, analytics provided through the SAS solutions for workforce analytics can then be used to thoroughly understand the relationship between work schedules/telework programs. This insight can help executives identify additional benefits that can be used to retain talent in an environment where traditional financial benefits, such as cash awards or promotions, are not possible due to shrinking budgets. SAS data integration tools can also be used to automate the process of making the millions of intelligent linkages among the various data sources – for example, highperforming employees with job titles, training regiments, job location, promotion potential and more. Agencies can also use SAS data integration tools to merge publicly available data from the Office of Personnel Management (OPM) across the federal government (and operating divisions within these agencies) to compare aspects of like agencies. For example, agencies can obtain benchmarks and compare the turnover rate of similar-size organizations or agencies with similar missions and job classifications.

6

Workforce Analytics for Government Agencies

SAS® Solutions for data cleansing and quality The SAS® Data Quality Solution provides an enterprise solution for profiling, cleansing, augmenting and integrating data to create consistent, reliable information. Agencies can automatically incorporate data quality into their data integration and business intelligence projects to dramatically improve “returns” on HR policy investigations and workforce what-if scenario analyses. For example, they can profile, monitor and actively manage the quality of data being collected from various sources, as well as integrate and standardize data across multiple systems. The software also allows agencies to define data correction rules to reflect organizational changes and cleanse data. This gives HR analysts trustworthy information needed to run analyses that empower executives to make informed decisions about how to manage their workforce today and the future.

Workforce Analytics: Turning Personnel Data into Information and Insight After agencies have a solid data foundation, they can apply analytics, such as data mining and predictive modeling, to identify patterns and trends in the behaviors of the workforce. It is important to note that no single analytical technique is capable of addressing all workforce challenges. Each of the analytical techniques discussed below is one piece of a larger solution. By leveraging the full complement of analytical capabilities in concert with one another, HR decision makers can have a complete understanding of their workforce environment and confidence that they can make wellinformed and well-calculated decisions that will address current workforce challenges, as well as decisions three, five and even 10 years down the road.

Data mining Data mining (sometimes called data discovery or knowledge discovery) is the process of analyzing data from different perspectives and summarizing it into useful information. Agencies can use this information to understand trends in the workforce, discover unseen behaviors of employees, and find hidden cause-and-effect relationships within the workforce. Technically, data mining is the process of finding correlations or patterns among dozens of fields in large databases. Data mining is primarily used by HR organizations with a strong personnel focus. It enables these organizations to determine key relationships among “internal” factors such as age, skill set or work demand, and “external” factors such as economic indicators and private sector competition. It also enables them to drill down into summary information to view detail transactional data. Data mining can also be helpful to HR departments seeking to identify the characteristics of their most successful employees; for example, information such as the universities attended by highly successful employees can help HR focus recruiting efforts on the most promising targets.

7

SAS White Paper

Using SAS® Solutions for data mining With SAS solutions for workforce analytics, agencies can employ multiple data mining techniques and methodologies including – but not limited to – clustering, decision trees, regression and neural networks. As a comprehensive workbench for data mining, it supports each step of the data mining process: identifying the most significant variables; developing models using the latest algorithms; validating the accuracy and fitness of the model(s); and generating a scored data set with predictive values that can be deployed into operational applications. HR organizations of government agencies gain: • Superior analytic data preparation techniques and an innovative set of algorithms to improve model performance and accuracy. • An easy-to-use interface that helps both business analysts and statisticians build better models, faster.

SAS solutions for workforce analytics follow the SEMMA process. The acronym SEMMA – Sample, Explore, Modify, Model, Assess – refers to the core process of conducting data mining. Beginning with a statistically representative sample of data, SEMMA makes

• An ability to derive and share insights to improve quality and precision of resulting decisions.

it easy to apply exploratory

• A broad set of tools to support an efficient, integrated data mining process.

techniques; select and

Predictive modeling

transform the most significant

Data mining and predictive analytics are often thought to refer the same thing. But they are actually quite different. Data mining is focused on the past and the present; it analyzes data that is currently available and identifies patterns in the data. In contrast, predictive analytics is about building a model that will help predict future behavior. Data mining is an event-driven analytics technique that allows users to identify the reasons or motives that led a certain event to happen. Predictive analytics, in contrast, is a technique that gives insights into what will happen next – and how it will affect an agency’s labor force. Think of predictive analytics as the “next step” after data mining has occurred. For example, once executives understand what influences the behavior and decisions of their workforce, they can predict what their workforce will look in the future.

predictive variables; model the

As this discussion suggests, gaining foresight means that agencies can’t stop their analysis at data mining and describing past events. They need to build upon this firstlevel analysis, maturing their analytical capabilities so they can predict future events in time to make decisions that shape future outcomes in positive ways. For example, when federal HR organizations can predict which employees are most likely to leave, they can implement programs that will help retain the most valuable employees and ultimately save the agency the expenses associated with replacing them.

8

statistical and visualization

variables to predict outcomes; and confirm a model’s accuracy.

Workforce Analytics for Government Agencies

As this example illustrates, predictive modeling can help agencies see into the future and take steps to proactively prevent problems or encourage success. This type of analysis can help agency executives answer questions such as: • Which employees are at the highest risk of voluntarily leaving the organization? • What are the reasons employees leave? • Which reasons have the most statistical significance to why employees leave? • What is the profile of employees most likely to leave? • What is the risk to the organization if employees leave? • Are the top performers leaving? If so, why? Not only can this analytical technique measure turnover, but it can also depict relationships between selected employee characteristics such as salary, educational level, skill, length of service, and time in position. Using this data, employees can be ranked based on their probability of voluntary termination within a specific time period. Once the behavioral characteristics of employees likely to leave are identified, agencies can accurately anticipate changes and adopt plans to deal with them; for instance, succession planning can improve for lower-level employees and retention strategies can be developed for elite employees.

Using SAS® Solutions for predictive modeling Predictive modeling enabled by SAS software draws on several related disciplines to give business users valid, credible forecasts of future events and insights needed to proactively apply strategic human capital initiatives to help meet agency objectives. For example, chief human capital officers (CHCOs) can:

With predictive modeling, chief human capital officers (CHCOs) can build on the data of yesterday and the information of today to forecast – and respond to – likely future events.

• Identify workforce trends and forecasting changes before they happen. • Quantify the bottom-line impact of HR processes or proposed policy changes. • Discover potential problems and unusual patterns before they materialize and adversely impact the agency. • Model voluntary turnover and performance abilities to proactively identify key talent for retention and/or leadership development. • Anticipate, forecast and predict changes in human capital resources – within the organization and in the changing economic environment. • Enable HR to clearly demonstrate its contribution to achieving agency and OPM goals. This is especially important since HR executives cannot have a proportionate level of influence in agency-level decisions and policies until they offer compelling and clear data that supports fact-based decision making.

9

SAS White Paper

Forecasting Because a large part of management is preparing for the future, it’s important to have a good understanding of what the future holds – and this is particularly true when making workforce-related decisions. The more accurate agencies are at predicting future needs and requirements, the more on target their decisions will be. Predicting the future requires accurate forecasting capabilities. Most agency executives believe that they are already doing forecasting. But in most cases, they are not actually using statistical models for this activity. Instead, they use a variety of guessing games to make future predictions. For example, they may take the attrition rate of last year and add a certain percentage increase based on past experience. This is often referred to as “judgmental forecasting,” and it can be extremely unreliable compared to true statistical modeling. Many executives also assume that forecasting and planning refer to the same thing. It’s true that both planning and forecasting deal with future behavior. But they are actually quite different. Planning addresses the question, “What should the future look like? We deal with constraints such as limited budgets, legislative restraints and limited resources. How can our agency achieve its goals under these constraints?” Planning is also an activity, where the decision maker is in charge and can influence future behavior by setting targets, allocating money in new ways and managing resources more effectively. Forecasting, on the other hand, answers the question, “What will the future look like?” It involves looking at past behavior (through data mining and predictive modeling) and coming up with an educated guess about what the future will most likely look like. As previously discussed, historical employee data contains a wealth of information that can be used in forecasting. In fact, the most commonly used forecasting techniques place weights on past observations (such as promotion frequency and types of awards) and combine them in a sophisticated way to estimate future behavior and outcomes. This gives HR analysts the freedom to, for example, place more weight on the recent past, if they believe that the recent past will have a bigger impact on the future than the more distant past. (To continue our prior scenario, they could, for instance, put more weight on the performance of employees within the last five years, when the agency introduced telework schedules.) Once analysts set up a robust, statistical forecasting model using historical and other data, they can search for meaningful patterns that can be used to support executive decision making. For example, they may identify hiring seasonality or other trends that can be used to extrapolate into the future. Robust, accurate forecasting models don’t take into consideration random fluctuations (also referred to as random noise) that can throw off predictions. Furthermore, analysts must avoid introducing bias – for example, by assuming that an anomaly is indicative of a trend. It may not be – and it’s risky to plan based on an anomaly. Analytics can help you identify these instances and incorrect assumptions, thus preventing poor planning and inaccurate forecasts.

10

Workforce Analytics for Government Agencies

Using SAS® Solutions for forecasting SAS solutions for workforce analytics provide quick and timely forecasts through a userfriendly graphical user interface (GUI). The solution automates the process of producing high-quality forecasts while giving analysts the ability to modify models interactively. This approach makes complex and multiyear forecasting processes manageable and allows analysts to focus their time on the most important forecasts. Automation also means fewer manual inputs, which reduces the chance that agency politics or personal agendas can contaminate forecasts. SAS solutions also deliver forecasts that reflect the realities of the labor force, helping executives plan future events with confidence. It automatically selects the business drivers or events that aid in the forecasting process from variables supplied to the system in the modeling process. As a result, forecasts can better reflect the intricacies of the agency. At the same time, it automatically builds the most appropriate model for a given set of data and delivers forecasts that are as accurate as can reasonably be expected (given the nature of the behavior(s) being forecast). Finally, SAS solutions for workforce analytics potentially improve forecasting performance across all agency divisions, bureaus and locations, at any level of aggregation. A virtually unlimited model repository makes it possible to create more appropriate forecasting models for a wider range of behaviors. In addition, planners can test what-if scenarios and determine how they are likely to affect the future demand of the workforce. Graphical displays make it easy to visualize employee promotions, attrition and unexpected events so executives can better forecast and plan future organizational activities. As this discussion illustrates, SAS solutions for workforce analytics are doing much more than simple projections of future trends (which is what the average spreadsheet product can do). Rather, they empower HR analysts to exploit all the strengths of SAS forecasting, without compromise. They also automate the process of statistical forecasting as much as possible, freeing time to focus on the more important forecasts or to try to improve models that cannot be automated (for example, “forecasting by exception”).

Simulation and optimization Optimization determines the best combination of resources within a given set of constraints. For example, an analyst can ascertain how to allocate SES or senior management positions (GS15/14) given certain budgetary or union constraints. The science behind optimization is not that complex. If a person had a piece of paper, a pencil and millions of hours to figure out all the staffing allocation possibilities, then it would be simply to find the one allocation scenario that would have the best benefit. However, analysts don’t have millions of hours to devote to running these calculations, which is why they need a robust solution to do the work for them and identify the best answer in seconds – not hours, days or weeks.

11

SAS White Paper

Even though the results of an optimization routine are the best allocations possible (given the constraints entered into the solution), it shouldn’t be considered the final answer until the results are properly reviewed. Human capital management is both an art and a science. SAS can help with the science and math, but human intervention is needed to create the art. For example, analysts may see that a constraint was not considered or that the results would not be able to hold up politically. In these cases, the results can be used as a baseline from which to work, or the analyst can go back into the solution and change the constraint.

Using SAS® Solutions for simulation and optimization SAS solutions for workforce analytics provide a powerful array of optimization, simulation and project scheduling techniques to identify which HR actions will produce the best results, while operating within resource limitations and tight restrictions. They enable government agencies to properly consider alternative actions and scenarios and determine the best allocation of resources and the best plans for accomplishing agency goals. This capability is particularly valuable, given the current economic state of the US federal government, as it is likely that agencies will be asked to not only do more with less, but also be asked to make staffing reductions and deep budget cuts. SAS optimization capabilities will allow HR analysts to input constraints (such as, not more than “X” total full-time employees or not more than “X” dollars for salaries and benefits) and determine the best allocation of resources given those constraints in a matter of seconds. An HR analyst could be challenged with figuring out how to reduce the workforce of an organization by 8 percent. The analyst could apply an 8 percent cut to every organization in the agency, but a more careful, data-driven and surgical way of making cuts would be to apply an optimization capability. In this case, analysts can use it and conclude that 4 percent of the cuts should come from “organization A” (4 percent less than 8 percent) and 12 percent of the cuts should come from “organization B” (4 percent more than 8 percent), thus averaging the cuts at 8 percent. At this point, the analyst may see that these results will not be received well politically (for example, perhaps organization B is in the district of the Chairman of the Appropriations Committee, and he or she would never endorse or support such a plan). In this case, the analyst could input a new constraint into the solution that says, “Do not take more than 1.5 percent of the cut from organization B”, and then run the optimization routine again. Within seconds, the solution would provide new results based on the new/added constraint – but this time, the results would be palatable to the chairman.

12

Workforce Analytics for Government Agencies

As this example illustrates, there are numerous advantages to using optimization capabilities. First, they allow agencies to identify the best answers to workforce planning problems using a wide range of operations research methods. SAS offers the broadest spectrum of operations research modeling and solution techniques available today and includes state-of-the-art methods for mathematical optimization. The depth of detail and realism in modeling capabilities, control of optimization, simulation and scheduling processes, and integrated approach to data access and information delivery enables HR organizations to identify and distribute the best answers to complex planning problems. Second, SAS solutions enable analysts to build models interactively and experiment with changes to underlying data. They can modify constraints or variables and experiment easily to see the effects of changes to underlying data. A specialized modeling language enables them to work transparently and directly with symbolic problem formulations. And for any given workforce scenario, they can choose from a set of proven solution methods that help analysts formulate models and solve problems intuitively and efficiently, regardless of whether they are linear, nonlinear or quadratic. Third, the SAS solutions enable the CHCO to plan, manage and track HR projects and schedules through a single integrated system. As a result, the CHCO can more effectively manage projects to meet deadlines within resource limitations, create backup plans in the event of possible changes, and have a solution at the ready to address unforeseen variations in resource availability. The software is designed to handle complicated situations such as legislative changes, budget cuts or unforeseen world events that may require quick action by the CHCO. And finally, the SAS solutions can be used to develop and analyze potential solutions to a given problem and determine the best possible answer. For senior HR executives who are called to testify as to why they made certain decisions or how a conclusion was reached, the system can generate justification-related information automatically, complete with a list of the possible scenarios considered and why the ultimate decision made was the best possible one given the information and the constraints known at the time.

13

SAS White Paper

SAS: Keeping You on the Cutting Edge of Decision Support Solutions Human capital is any government agency’s best asset. So it’s critical that executives be able to organize and prepare their workforce in response to changes in organizational missions and the political and fiscal environments of the federal government. SAS is uniquely positioned to team with federal agencies to make this happen. SAS solutions for workforce analytics provide an end-to-end solution for gathering data, finding and understanding behaviors of the workforce, forecasting the supply and demand of the workforce, performing scenario analysis and optimizing resource allocations. Only SAS combines all of the analytical approaches outlined in this paper in an integrated set of solutions, empowering agencies to organize a more effective and efficient labor force for today and tomorrow. Furthermore, SAS is universally recognized as the worldwide leader of advanced analytics. SAS’ market share in predictive modeling alone is more than double our closest competitor’s. Only SAS can provide federal agencies with an open, highperformance and scalable solution for leveraging analytics throughout your entire workforce strategy – from right-sizing to the current government environment to planning the workforce of tomorrow. To learn more, visit sas.com/government.

14

About SAS SAS is the leader in business analytics software and services, and the largest independent vendor in the business intelligence market. Through innovative solutions, SAS helps customers at more than 60,000 sites improve performance and deliver value by making better decisions faster. Since 1976 SAS has been giving customers around the world THE POWER TO KNOW ®. For more information on SAS® Business Analytics software and services, visit sas.com.

SAS Institute Inc. World Headquarters   +1 919 677 8000 To contact your local SAS office, please visit:

sas.com/offices

SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration. Other brand and product names are trademarks of their respective companies. Copyright © 2013, SAS Institute Inc. All rights reserved. 106165_S94845_0113