DATAWorks Speakers and Abstracts

Dr. James Wisnowski

Adsurgo
Design and Analysis of Experiments – Next-Level Methods with Case Studies
Day 1, Room: A 9:00 AM-4:00 PM

Dr. James Wisnowski, co-founder and principal consultant at Adsurgo, leads the enterprise consulting and government divisions of Adsurgo. Dr. Wisnowski has consulting experience and expertise in applied statistics, program management, strategic planning, military operations, design of experiments, reliability engineering, quality engineering, data mining, text analytics, simulation modelling, along with operations research analysis. He has published refereed journal articles and texts in addition to presenting consulting results, new research, and short courses at conferences worldwide. He retired from the US Air Force as an officer with 20 years of service as an acquisition, test, personnel, and force structure analyst in addition to having significant leadership responsibilities as a squadron commander, joint staff officer, and Air Force Academy professor.

Abstract: Design and Analysis of Experiments – Next-Level Methods with Case Studies

This is the short course for you if you are familiar with the fundamental techniques in the science of test and want to learn useful, real-world, and advanced methods applicable in the DoD/NASA test community. The focus will be on use cases not typically covered in most short courses. JMP software will primarily be used, and datasets will be provided for you to follow along many of the hands-on demonstrations of practical case studies. Design topics will include custom design of experiments tips, choosing optimality criteria, creating designs from existing runs, augmenting adaptively in high gradient regions, creating designs with constraints, repairing broken designs, mixture design intricacies, modern screening designs, designs for computer simulation, accelerated life test, and measurement system testing. Analysis topics will include ordinary least squares, stepwise, and logistic regression, generalized regression (LASSO, ridge, elastic net), model averaging (to include Self-Validated Ensemble Models), random effects (split-plot, repeated measures), comparability/equivalence, functional data analysis (think your data is a curve), nonlinear approaches and multiple response optimization and trade-space analysis. The day will finish with an hour-long Q&A session to help solve your specific T&E challenges.