Home / Courses / Training / Design of Experiments (DOE) for Method Validation & Laboratory Studies

Design of Experiments (DOE) for Method Validation & Laboratory Studies

This one-day, application-based workshop introduces the principles and practical tools of design of experiments (DOE) for calibration, testing, and other measurement-based laboratories. Participants learn how to convert a technical question into an experimental objective, select responses and factors, establish factor levels and ranges, and build an experimental plan that produces interpretable data. Emphasis is placed on two-level factorial designs; randomization, replication, blocking, and center points; main effects and interactions; analysis of variance (ANOVA); residual review; and defensible technical conclusions. Using a provided Microsoft Excel workbook and laboratory case studies, participants will plan and analyze experiments for method development and robustness, process improvement, competence studies, measurement system investigations, and evaluations of contributors to measurement uncertainty. 

Prerequisites

Required: Familiarity with basic algebra and descriptive statistics, including the mean, standard deviation, and variance. Participants should be comfortable entering formulas, sorting data, and creating basic charts in Microsoft Excel or equivalent software. 

Desired: Prior exposure to confidence intervals, hypothesis tests, and one-way ANOVA; experience performing calibration, testing, method validation, or engineering studies; and a current laboratory question that could benefit from a structured experiment. 

Equipment Needed

In-Person (on Location)
Laptop (PC or Mac)
Microsoft Excel 2016/365 or equivalent; Analysis ToolPak enabled when available
Provided workbook, data files, and DOE planning worksheet
Calculator or calculator application

Virtual Delivery
Laptop (PC or Mac)
Microsoft Excel 2016/365 or equivalent; Analysis ToolPak enabled when available
Provided workbook, data files, and DOE planning worksheet
Reliable internet connection, headset, and calculator application; second monitor recommended

Target Attendees

This workshop is intended for personnel who plan, perform, review, or approve laboratory and engineering studies and who want to obtain more information from experimental data than is typically available from one-factor-at-a-time trials. 

Organizational Level
Directors, Managers, and Supervisors
Quality and Technical Managers
R&D Scientists, Method Developers, Engineers, and Metrologists
Laboratory Analysts and Technicians

Learning Outcomes

After successful completion of the workshop, participants will be able to, using a provided Excel workbook and laboratory case studies: 

  1. Translate a laboratory or engineering problem into a clear experimental objective, response, and decision statement.
  2. Identify controllable factors, noise factors, practical factor ranges, levels, constraints, and potential responses.
  3. Distinguish one-factor-at-a-time trials from designed experiments and select a screening, characterization, or optimization strategy.
  4. Apply randomization, replication, blocking, balance, and center points to reduce bias and support valid conclusions.
  5. Construct and interpret a two-level full factorial design, including coded settings, main effects, and two-factor interactions.
  6. Explain fractional factorial designs, design resolution, aliasing, and confounding at an introductory level.
  7. Use the supplied Excel workbook to estimate effects, review ANOVA output, create interaction plots, and examine residuals.
  8. Recognize common DOE pitfalls, including inadequate measurement capability, uncontrolled factors, pseudo-replication, missing runs, and over interpretation of p-values.
  9. Prepare a concise DOE plan and results summary that records the design, execution, analysis, conclusions, limitations, and confirmation

CEUs Awarded

0.7 CEUs Awarded (7 contact hours, 8 hours total) 

Presentation Style

Short lectures, guided discussions, instructor demonstrations, Excel-based exercises, and a capstone laboratory case study. Participants receive a DOE planning worksheet, an analysis workbook, example data sets, and a concise course reference guide. 

Reference

This course follows the principles and terminology outlined in the following documents and references: 

  • ISO 3534-3, Statistics – Vocabulary and symbols – Part 3: Design of experiments; Common terminology and concepts for designed experiments.
  • NIST/SEMATECH e-Handbook of Statistical Methods, Chapter 5: Process Improvement; Practical guidance and examples for factorial designs, analysis, and process improvement.
  • ISO/IEC 17025:2017; Laboratory context for method selection and validation, technical records, control of data, and validity of results.
  • Eurachem, The Fitness for Purpose of Analytical Methods: A Laboratory Guide to Method Validation and Related Topics; Application of experimental studies to method performance, robustness, validation, and fitness for purpose.
  • Douglas C. Montgomery, Design and Analysis of Experiments; Design selection, factorial methods, ANOVA, diagnostics, and sequential experimentation.
  • George E. P. Box, J. Stuart Hunter, and William G. Hunter, Statistics for Experimenters; Practical experimental strategy, discovery, interactions, and iterative learning.