Summary

Data are a strategic asset. The effective use of data science—the science and technology of extracting value from data—improves, enhances, and strengthens acquisition decision-making and outcomes. Those outcomes include improvements in missions, portfolios of capabilities, and the schedules, costs, and performance of acquired systems for net improvements to warfighters, operators, and supporting organizations. Using data science to support decision making is not new to the defense acquisition community; its use by the acquisition workforce has enabled acquisition and thus defense successes for decades. Still, more consistent and expanded application of data science will continue improving acquisition outcomes, and doing so requires coordinated efforts across the defense acquisition system and its related communities and stakeholders. Central to that effort is the development, growth, and sustainment of data science capabilities across the acquisition workforce.

The Under Secretary of Defense for Acquisition and Sustainment [USD(A&S)] tasked the Committee on Improving Defense Acquisition Workforce Capability in Data Use to assess how data science can improve acquisition processes and develop a framework for training and educating the defense acquisition workforce to better exploit the application of data science. Key to the framework’s development was the identification of opportunities where data science can improve acquisition processes, the relevant data science skills and capabilities necessary for the acquisition workforce, and relevant models of data science training and education.

CONCLUSIONS AND RECOMMENDATIONS

Continue Improving Defense Acquisition with Data Science

There are examples of the successful application of data science in Department of Defense (DoD) acquisition. There are also clear opportunities to better use data science across acquisition policy making, processes, decision making, and actions. Data science has improved acquisitions processes, for example, by enhancing program and contracting analysis, tracking cost and program performance, enabling analysis of alternatives, determining whether a weapon or platform meets performance requirements, and informing decision making. Typically, improvements can be made in the collecting, curating, and managing phases of the data life cycle for broad range of acquisition functions, including cost estimating, contracting, contract management and cybersecurity. Currently, there are several shortcomings in the front end of data science processes in defense acquisition—namely in data collection and curation. Also, data silos can hinder shared data across DoD and are a common obstacle preventing the utilization of the full potential of the data life cycle. The vision and opportunity for data science in the defense acquisition system is one in which data collection and use is not a support function but is integrated into all acquisition processes.

The Secretary of Defense and other senior leaders can facilitate adoption by demanding high-quality, complete, and accurate data for their decisions and those below them while providing the necessary financial and personnel resources for obtaining these data and analyses. This pull is necessary to drive behavior and complement the push from technical opportunities, putting action behind their strategic statements.

Opportunities Exist to Improve Workforce Abilities Across the Data Science Life Cycle

Extracting value from data requires a collective data science mindset, skillset, and toolset. Central to data science is the data life cycle, which is an iterative, bi-directional workflow—with people at each element—that has several phases: questions, definitions, coordination, generation of data, collection of data, curation and management of data, analysis of data, dissemination and interpretation, and assessment.

The committee found that an increasing majority of the acquisition workforce is participating in the data life cycle but may not recognize their roles and value within it. Understanding their roles across the data life cycle is critical for improved and sustained data use and data-informed decision-making. For utilizing the full potential of the data life cycle, teams will need to be established, customized, nurtured, and managed. Team leaders will also need familiarity with the data life cycle, management skills that support and optimize data science talent, and a commitment to data-informed decision-making. Managing data science projects uses strategies and approaches for leading collaborative, cross-functional, technical projects. Specific attention is necessary for the development of a team with skills across the data life cycle and the ability to ask questions specific to the quality and utility of data.

Data Literacy Skills Are Important Throughout the Acquisition Workforce

DoD and its components should ensure that all members of the acquisition workforce have at least basic (non-technical) data science skills, which include an understanding of the data life cycle and how it works, an ability to consume and understand data story-telling and data-informed decisions, and an ability to recognize matters of ethics, privacy, and security. These skills make up what is often called data literacy—a set of data-centric skills that have evolved with data science and are growing in importance and use in government, industry, and academia.

Advanced Capabilities Can Enable Improved Data Use in Acquisition

Collectively, the defense acquisition workforce must have capabilities in all phases of the data life cycle—from data collection and curation to data analysis and visualization—to fully exploit the value of a particular data stream and its corresponding domain application. Executing the data life cycle is a collaborative endeavor, requiring a collective skillset found in teams of data engineers, data scientists, data analysts, data users, domain experts, and leaders/decision makers. Data scientists are experts across the data life cycle, with special emphasis on advanced techniques for curation, management, analysis, visualization, dissemination, and interpretation. The vast majority of the defense acquisition workforce will identify as domain experts, leaders/decision makers, and even data analysts and data users (under the guise of their acquisition job functions), but data engineers and data scientists are also critical for improved data use.

Diverse, Tailored, and Situated Training Models Can Increase Data Capabilities and Outcomes in the Acquisition

Workforce development in data science is critical to the success of DoD’s current and future acquisition improvement efforts. The committee examined a variety of training efforts within and outside of DoD for increasing data capabilities of the defense acquisition workforce. Currently, there are several non-DoD data science training efforts that could be leveraged for the acquisition workforce; some are already being used, but use could be expanded. To be effective, data science training needs to be applied in the context of acquisition functions—preferably with realistic data and examples—rather than simply taught in the abstract without applications. Also, there is no one-size-fits-all training approach; training must be tailored to meet specific needs depending on professional roles, the application and scale of data, and mission goals and priorities. In addition, as data science curricula and courses are piloted for the acquisition community, clear metrics are key for evaluating the success and applicability of the training

approaches. Successful training efforts can be scaled to increase impact and broaden the data capabilities of the defense acquisition workforce.