Research and development at NASA is a process of discovery and learning as the scientific method is employed to iteratively move between inductive and deductive reasoning as described by Box (1999). While the discovery process can take many different paths, the amount and rate of knowledge obtained with the available resources is a measure of effectiveness. Strategic application of statistical methods is essential to make this process efficient and rigorous. In this presentation, we provide an overview of statistical practice and applications from a statistical engineering team located at NASA’s Langley Research Center. Our team focuses on engineering statistical sciences to generate improved solution approaches for a wide variety of applications across NASA’s missions of aeronautics, space exploration, and atmospheric science. The team’s mode of statistical practice is characterized by close collaboration with other science and engineering disciplines, a synergistic combination of statistical methods, and solutions that are embedded into standard work processes. In addition, problem-driven research is conducted to adapt and extend statistical methodologies, particularly in statistical design of experiments and response surface methodology. We discuss the structure of the team, including the broad discipline training and experience, along with active external collaboration. To illustrate typical statistical challenges we encounter, three case studies are presented from estimating solid rocket motor performance, characterizing a Mars entry measurement system, and studying how NASA satellites and data products could be used to enhance soybean production. All of these examples emphasize the statistician’s role in defining the right questions and implementing multiple statistical methods to meet high-level objectives.

Box, G.E.P. (1999) “Statistics as a Catalyst to Learning by Scientific Method Part II - A Discussion,” Journal of Quality Technology, 31 (1), pp. 16-29.

Short biographical note: Dr. Peter Parker is a member of the Aeronautics Systems Engineering Branch in the Engineering Directorate at NASA’s Langley Research Center. At Langley, Pete leads the Statistical Engineering Team that works across all of NASA’s mission directorates to infuse statistical thinking, engineering, and methods. His particular areas of expertise include statistical design of experiments, response surface methodology, and measurement system characterization. He holds a B.S. in Engineering, a M.S. in Applied Physics and Computer Science and a M.S. and Ph.D. in Statistics from Virginia Tech. Pete is a senior member of the American Society for Quality, American Statistical Association, and the American Institute of Aeronautics and Astronautics. He currently serves as a Senior Editor and is Editor-elect for the American Society for Quality’s journal Quality Engineering. Pete is the recipient of numerous awards, including the NASA Office of Chief Engineer’s Systems Engineering Activity of the Year under Techniques and Methodologies for his application of design of experiments to the Shuttle flow control valve poppet anomaly and most recently he was named as an Agency recipient of the Federal Engineer of the Year Award by the National Society of Professional Engineers.