2026 FTC Short Courses

Short courses will be held on October 6, 2026. Attendees may registers for the conference and short courses together or separately. Conference attendance is not required to attend a short course.

Scott Mutchler – From Prompt to Deployment: Using Coding Assistants to Generate Reproducible Quarto Reports for Quality Data Analysis

Peter Goos – OMARS Designs: Modern Experimental Designs for Simultaneous Screening and Response Surface Optimization

MORNING HALF DAY COURSE

From Prompt to Deployment: Using Coding Assistants to Generate Reproducible Quarto Reports for Quality Data Analysis 8:00am to 12:00pm

Scott Mutchler, Virginia Tech

Abstract: The integration of large language model (LLM)-assisted coding tools into statistical workflows presents both significant opportunity and methodological risk for quality practitioners. This course demonstrates a practical framework for leveraging coding assistants to scaffold, generate, and validate reproducible Quarto reports for data analysis in quality applications, including process capability studies, control chart interpretation, measurement system analysis, and designed experiments. Hands on exercises will be provided for attendees. A central challenge in AI-assisted analysis is ensuring that generated code is not merely functional but auditable, statistically defensible, and reproducible across computing environments. We present a structured prompting and review workflow in which coding assistants generate fully documented Quarto documents (.qmd) with embedded R or Python code, interpretive narrative, and diagnostic visualizations—while the analyst retains explicit control over analytical decisions and output validation. Key design principles include version-controlled project structures, clearly scoped system prompts that encode domain conventions (e.g., Shewhart rules, AIAG MSA guidelines), and human-in-the-loop checkpoints for assumption verification. Case examples drawn from manufacturing and service quality contexts illustrate how this approach reduces time-to-report while maintaining—and in some cases improving—analytical transparency relative to traditional point-and-click workflows. We also address failure modes observed in practice: hallucinated package functions, statistically inappropriate defaults, and narrative overconfidence, along with mitigation strategies. Attendees will leave with a reproducible template workflow and practical guidance for evaluating AI-generated statistical code before it reaches a client or decision-maker.

Target audience: Anyone that wants to leverage agentic AI tools for analysis and/or automation.

Prerequisites: Basic computer skills. This session will be hands-on with Claude Code, Python and Quarto. Participants will be required to pre-install these on their PCs. The instructor will provide detailed instructions and an API key for participants to use Claude Code (without cost).

Contact: smutchler@vt.edu

Scott Mutchler is an associate professor of practice in the Academy of Data Science, brings over 25 years of experience in advanced analytics, machine learning, and enterprise software development, with a special passion for driving competitive advantage through generative AI and optimization technologies.
Throughout his career, Mutchler held senior leadership positions at major consulting and technology firms, where he established successful analytics practices and advised global executives on strategic analytics implementations. His innovative work spanned manufacturing, distribution, and retail sectors, including the development of generative AI applications, manufacturing optimization tools, and IoT-based predictive maintenance solutions. His work in automated retail planning systems earned industry recognition, with his solutions generating significant ROI for multiple Fortune 500 clients.

A dedicated educator and mentor, Mutchler has served as an adjunct IT professor in addition to his current role at Virginia Tech. He holds multiple patents in retail analytics and maintenance optimization and has authored technical publications on data mining and analytics. Mutchler earned his M.S. in Geology from Virginia Tech and his B.S. in Geology from Indiana University of Pennsylvania.

AFTERNOON HALF DAY COURSE

OMARS Designs: Modern Experimental Designs for Simultaneous Screening and Response Surface Optimization 1:00pm to 5:00pm

Peter Goos, University of Leuven

Abstract:  Traditionally, factor screening and response surface experimentation have always been performed in two separate steps. The increasing pressure to speed up innovation necessitates an adaptation of that strategy, by combining factor screening and optimization into a single experiment. In recent years, suitable economical experimental designs have been developed to study many factors simultaneously. These designs are called Orthogonal Minimally Aliased Response Surface or OMARS designs. The family of OMARS designs forms a major extension of the family of definitive screening designs. Due to the fact that they involve three levels per factor, their larger power for detecting quadratic effects and their good projection properties, OMARS designs serve as screening experiment and as response surface experiments at the same time. In this short course, we first discuss the properties of OMARS designs, as well as various design quality characteristics that can be used to select OMARS designs that are appropriate for a given problem. A key aspect in our design selection is the use of multiple criteria to select a good experimental design, rather than a single design optimality criterion. We also discuss extensions of the family of OMARS designs. For instance, we introduce mixed-level OMARS designs which allow the study of two-level factors in addition to three-level factors. We also explain how OMARS designs can be arranged in blocks, so that the resulting designs are orthogonally blocked. Finally, we introduce the concept of strong OMARS designs, which are economical orthogonal designs with the same orthogonality properties as central composite and Box-Behnken designs, but with fewer runs and a higher information content. Throughout the course, we will discuss algorithmic and combinatorial constructions, applications and availability of OMARS designs.

Target audience: Practitioners facing complex, expensive experiments as well as statisticians who are curious about modern design of experiments.

Prerequisites: Solid knowledge about classical design of experiments (screening, response surface optimization) and analysis of experimental data (linear regression analysis).

Contact: peter.goos@kuleuven.be

Peter Goos is a full professor at the Faculty of Bio-Science Engineering of the University of Leuven and at the Faculty of Business and Economics of the University of Antwerp, where he teaches various introductory courses on statistics and probability. Since 2026, Peter is also a thought leader and consultant for Minitab. His main research area is the statistical design and analysis of experiments. Besides a number of articles in top scientific journals in marketing, transportation, quality, operations research and statistics, he published the books “The Optimal Design of Blocked and Split-Plot Experiments” and “Optimal Experimental Design: A Case-Study Approach”. For his work, Peter Goos has received the Brumbaugh Award, the Youden Award, the Shewell Award and the Lloyd S. Nelson Award of the American Society for Quality, the Ziegel Award and the Statistics in Chemistry Award from the American Statistical Association, and the Young Statistician Award of the European Network for Business and Industrial Statistics.