The statistic experimental design (English Design OF of experiment, DOE) is used during development and optimization by products or processes.

Since attempts need resources (personnel, time, devices etc.) see itself the attempt responsible person in a Zwiespalt between accuracy and reliability of its expected results and the expenditure necessary for it.

With the statistic experimental design with a minimization of the number of attempts the factors of influence and reciprocal effects are determined on the goal sizes. Also the size of the effects is determined, whereby an optimization of the processes which can be examined becomes possible.

Attempt plans

So-called attempt plans are provided, which consider the following:

• Number of factors which can be examined (at least 2)
• Kind of the factors which can be examined (nominally, ordinal, continuously)
• Existing information
• Desired accuracy/reliability of the statements

Classical plans in the experimental design are factorial and part-factorial plans (so-called Screening Designs), the latters offer the possibility of a strong reduction of the number of attempts. Results from Screening Designs can be transferred also to a later attempt series with less examined factors to Response Surface plans.

Response Surface plans serve for to the determination and following optimization of nonlinear connections.

Evaluation

In the long run the result is a mathematical model of the received attempt data, which exhibits a certain statistic security. In principle the following involution models stand to the selection:

• Fringe effects: Linear model without reciprocal effects
• Main and side effects: Linear model with reciprocal effects
• Main and side effects: Square (rarely cubic) models with reciprocal effects

Related links

• Experimental design: Multi-factor experiments
• Freeware for experimental design with Excel

Articles in category "Statistic experimental design"

We found here 10 articles.

Related Websites

We found here 6 related websites.

• EXPERIMENTAL DESIGN
  Then we not only need statistical tests to analyse those differences but we also need good experimental design to ensure that we haven't biased our results ...


• Homepage of Chemometrics
  In 1925 Fisher started the development of methods of statistical experimental design. These methods have been further refined by Yule, Box, ...


• SFU Statistics & Actuarial Science - Stat 650/403
  Learn how to design and analyze some of the common experimental and survey designs. Enjoy the subject of statistics rather than dreading it. ...


• Statistics Glossary - design of experiments & ANOVA
  statistical glossary - design of experiments & ANOVA. ... Experimental Design. We are concerned with the analysis of data generated from an experiment. ...


• Statistics Glossary - Design of Experiments & ANOVA
  The structure of the experiment in a completely randomised design is assumed to be such that the treatments are allocated to the experimental units ...


• Wiley::Statistics Experimental Design
  Experimental Designs, 2nd Edition by William G. Cochran, Gertrude M. Cox Statistics for Experimenters: Design, Innovation, and Discovery, 2nd Edition ...