2022 · Statistics
Design of Experiments (DOE) Analysis
Statistics DOE R
Design of Experiments (DOE) Analysis
BITS Pilani | Sep 2022 – Dec 2022
R Factorial Design Response Surface ANOVA Process Optimization
Project Overview
Applied statistical design of experiments methodology to systematically optimize industrial processes, identifying optimal factor settings and interaction effects.
Key Contributions
Factorial Analysis: Designed and analyzed 2^k full factorial and fractional factorial experiments to identify significant main effects and interactions among process variables
Response Surface Optimization: Applied Response Surface Methodology (RSM) with Central Composite Design (CCD) to locate optimal operating conditions, improving process efficiency by 25%
Statistical Inference: Conducted ANOVA, regression diagnostics, and lack-of-fit tests to validate model adequacy and identify statistically significant effects (α = 0.05)
Technologies Used
- Languages: R
- Libraries: rsm, FrF2, DoE.base
- Methods: Factorial Design, RSM, ANOVA
- Visualization: contour plots, interaction plots
Key Results
| Metric | Value |
|---|---|
| Process Improvement | 25% |
| Design Type | CCD, 2^k Factorial |
| Significance Level | α = 0.05 |
| Factors Analyzed | Multiple |
DOE Workflow
1. Problem Definition → Response Variable Selection
2. Factor Identification → Screening Experiments
3. Full Factorial Design → Main Effects & Interactions
4. Response Surface Modeling → Optimization
5. Confirmation Runs → Validation