Application of experimental design to generate relevant information and representative calibration data
1997

Using Experimental Design for Calibration Data

Sample size: 16 publication Evidence: moderate

Author Information

Author(s): Suzanne Schönkopf, Dominique Guyot

Primary Institution: Camo ASA

Hypothesis

Can experimental design techniques improve the selection of representative calibration data?

Conclusion

The study demonstrates that using experimental design can effectively generate representative calibration data for predictive modeling.

Supporting Evidence

  • Experimental design can create structured variation in data.
  • Fractional factorial designs require fewer experiments while still providing balanced data.
  • Using representative data improves the chances of successful calibration models.

Takeaway

This study shows that by carefully choosing how to experiment, we can get better data to help us make predictions.

Methodology

The study utilized factorial and fractional factorial designs to systematically vary experimental conditions and identify significant factors affecting yield.

Potential Biases

The selection of samples based on experimental design may introduce bias if not all variations are adequately represented.

Limitations

Some experiments could not be performed with the stipulated settings, which may affect the analysis.

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication