Partial factor plan

In the statistical experimental design is meant by a fractional factorial design (ger .: fractional factorial design ) an experimental design that only a certain fraction of a full study plan by playing factor combinations contained.

For example, a full design for six factors at two levels would require trial runs. For reasons of economy, a partial factor plan of the type or etc. is often used in such a case . The notation here indicates that only a fraction of the total possible test runs, namely , is carried out and thus test runs. A so-called orthogonal field is often used to select the eight test runs contained in the partial factor plan . ${\displaystyle 2^{6}=64}$ ${\displaystyle 2^{6-3}}$ ${\displaystyle 2^{7-4}}$ ${\displaystyle 2^{6-3}}$ ${\displaystyle 2^{6}}$ ${\displaystyle 2^{-3}={\frac {1}{8}}}$ ${\displaystyle 2^{6-3}={\frac {64}{8}}=8}$

The disadvantage of partial factor plans is the thereby occurring blending of factor effects and interactions . This is all the more noticeable the smaller the selected fraction of the complete test plan. To describe the type and quantity of the mix-ups that occur, each partial factor plan is assigned a resolution class (English: resolution number ).

Individual evidence

1. ^ ^{A b} Montgomery, Douglas C .: Design and Analysis of Experiments. John Wiley and Sons, 1991, ISBN 0-471-52994-X , p. 335
2. ^ Montgomery, Douglas C .: Design and Analysis of Experiments. John Wiley and Sons, 1991, ISBN 0-471-52994-X , p. 360
3. ^ Montgomery, Douglas C .: Design and Analysis of Experiments. John Wiley and Sons, 1991, ISBN 0-471-52994-X , p. 418
4. ^ Montgomery, Douglas C .: Design and Analysis of Experiments. John Wiley and Sons, 1991, ISBN 0-471-52994-X , p. 339

Web links

• Fractional Factorial Designs (National Institute of Standards and Technology)