STAT Center of Excellence

Reliability and Reliability Growth (REL)

REL 220:  Reliability for Practitioners

COE-R 310: Reliability and Reliability Growth Foundations I This four-day course covers the policies, guidance, and methods to improve the reliability program across the acquisition life cycle. Focus is on both the proactive approach of designing reliability into the system up front (Design for Reliability) and the reactive Reliability Growth modeling. Emphasis is on understanding basic reliability and Reliability Growth principles and using tools available to the practitioner to plan, assess, track, and project reliability. This hands-on course uses software to reinforce analytical methods presented with demonstrations and exercises. Students will be able to construct the Reliability Growth curves now required in Test and Evaluation Master Plans for major systems. Prerequisites: COE-R 210 or REL 210.

COE-R 410: Reliability and Reliability Growth Foundations II  This four-day course is intended for AF personnel engaged in system lifecycle management who require an advanced application level understanding of reliability and reliability growth. This course reinforces the fundamentals from REL 310 and delves more deeply into both design and sustainment reliability analysis techniques while assessing the impact of reliability across the life cycle. Focus is on both the proactive approach of designing reliability into the system up front (Design for Reliability) and the reactive Reliability Growth modeling. Prerequisites: COE-R 310 or REL 310.

Design of Experiments (DOE)

DOE 320: Design of Experiments for Managers

This is a two-day introduction to the Science of Test with emphasis on Design on Experiments (DOE). Though ‘experiments’ is in the name, these methods apply to any test program throughout the entire test life cycle: planning, design, execution, and analysis. This course is intended for practitioners, managers, and stakeholders as an exposure to the basic concepts of Scientific Test and Analysis Techniques (STAT) and how they can be applied to test programs. This course covers the same topics as the one week course (DOE 252) but with less depth and no practical exercises.

DOE 251: Design of Experiments for Practitioners

This five day course in designed tests teaches students basic techniques and processes needed to create a statistically rigorous and defensible test for complex systems. A well-designed experiment can lead to reduced development lead time with fewer tests required, provide greater insight into system performance, and ultimately lead to fielding better, more reliable systems. Students will learn how to plan, conduct and analyze tests efficiently. Using examples from military T&E and the students’ own test experiences, the course has a strong applied flavor. Students will learn a disciplined approach to clearly define the true test objectives, measures of effectiveness (MOE)/performance (MOP), appropriate input factors that impact the MOEs/MOPs, and the recommended settings for these factors to effectively span the battlespace. Students will design powerful tests through appropriate sample size computations, construct efficient test run matrices, develop statistical models, test models for assumptions, and quantify the uncertainty of test results. Factorial, fractional factorial, and custom designs are the methods used throughout this course. In class projects and exercises will reinforce learning objectives.

DOE 252: Design of Experiments for Practitioners

This course reinforces the fundamentals from DOE 251, examines (more closely) what makes good designs across a variety of metrics, introduces new classes of designs, and provides advanced modeling and analysis methods. Design evaluation topics include power, sample size, optimality, and aliasing criteria. Students will learn additional design approaches for realistic test issues such as constrained test regions, multiple factor levels, categorical factors, nuisance factors, and hard-to-change factors. Sequential design is highlighted with augmentation methods. The course begins by reinforcing and building upon student knowledge (from DOE 252) on factorials, power, fractional factorials, mixed level, and response surface designs. Advanced material on fractional factorial designs for screening are addressed with additional focus on response surface methods to model nonlinear behavior particularly when the test objective is to optimize performance; several variations of response surface designs are presented. The course concludes with a survey of advanced DOE topics tailored to the organization’s requirements including multiple response optimization, design optimality (with an introduction to robust screening and alias optimal designs), and split-plot designs for hard-to-change factors. Prerequisites: DOE 251

Scientific Test and Analysis Techniques (STAT)