Design of Experiments¶
The DOE (Design of Experiments) module provides a comprehensive framework for creating experimental designs that help you efficiently explore factor effects and interactions while minimizing the number of experimental runs required.
Overview¶
Design of Experiments is a systematic approach to understanding how different factors (variables) affect a response. Instead of changing one factor at a time, DOE allows you to study multiple factors simultaneously, revealing interactions between factors and providing more information with fewer experiments.
Key Features¶
- Multiple Design Types: Support for full factorial, fractional factorial, and specialized 2^k designs
- Flexible Factor Definition: Handle both numerical and categorical factors with custom levels
- Advanced Options: Replication, blocking, central points, and randomization
- Foldover Support: Resolve aliasing in fractional factorial designs
- Automatic Encoding: Smart conversion between original factor values and coded levels
When to Use DOE¶
- Process Optimization: Find optimal settings for manufacturing processes
- Product Development: Understand how design parameters affect performance
- Quality Improvement: Identify factors that affect product quality
- Screening Studies: Determine which factors are most important
- Response Surface Methodology: Build predictive models of your system
Quick Start¶
Here's a simple example of creating a full factorial design:
import daspi as dsp
# Define factors
temperature = dsp.Factor('Temperature', (150, 200))
pressure = dsp.Factor('Pressure', (10, 15))
catalyst = dsp.Factor('Catalyst', ('A', 'B'), is_categorical=True)
# Create design
builder = dsp.FullFactorialDesignBuilder(
temperature, pressure, catalyst,
replicates=2,
shuffle=True
)
# Generate the experimental design
design = builder.build_design(corrected=False)
print(design)
This creates a 2×2×2 factorial design with 2 replicates (16 total runs), showing all possible combinations of the three factors.
Documentation Structure¶
The DOE module documentation is organized into the following sections:
- Factor: Learn how to define experimental factors with their levels
- Design Builders: Understand the base functionality and common features
- Full Factorial Designs: Explore complete factorial designs and 2^k designs
- Fractional Factorial Designs: Discover reduced designs with foldover support
For practical guidance and real-world examples, see the DOE User Guide.