d e f es 
AICOS News
15.03.2020
Coronavirus: continuation of our activities
Independently of the coronavirus epidemic... 
more Linie
20.02.2020
AICOS Technologies at PCH Meetings 2020 in Lyon
Lyon - 25.-26.11.2020: International business... 
more Linie
05.01.2020
Our training schedule for spring / summer 2020 is on-line!
You want to deepen your knowledge in data... 
more Linie
Search
Go
Infos > Design of Experiments (DoE)

Statistical Design of Experiments (DoE) in practice

Most effort is needed before performing the experiments

The systematic approach of statistical Design of Experiments (DoE) forces the user to clearly define the questions which are to be answered in the project.
This can be a very time-consuming task, as it has to be specified:

  • which quality criteria ("response variables ") to investigate,
  • whether only a few important response variables should be included, or whether the opportunity should be used to investigate some additional parameters,
  • what desired optimal outcome is aimed at (for each response),
  • which are the potential influence factors,
  • which region should be chosen for each of the influence factor (not too narrow and not too wide),
  • whether some factors are to be fixed in order to save time, and if yes, at which value (knowing the risk that for some other setting of this factor, an altogether better result could be reached),
  • which number of experiments is feasible,
  • whether there is prior knowledge about interactions,
  • whether interactions should be included into the investigation or whether they should be neglected for the moment,
  • ...

Then everything runs smoothly

After the research question has been clarified, working with a user-friendly DoE tool such as e.g. STAVEX is very intuitive.
You enter the quality criteria of interest, their optimisation direction, as well as the potential influence factor and their regions of variation. STAVEX then suggests a suitable experimental design. bilder/ProductsSTAVEX/STAVEX_4d-plot-e.jpg . This means for instance, that when the number of potential influence factors is large, one should not expect very detailed information. Otherwise the number of experiments would become too large (key word sequential Design of Experiments).
You perform the experiments, then STAVEX assists you with the statistical analysis and its interpretation. The rich graphics library allows for a comprehensive visualisation of the results.

After a single experimental cycle the project is not necessarily finished

The decision for the next step again needs some consideration and perhaps exchange with colleagues. However, also here you can count on the best possible assistance of STAVEX.

Also complex DoE problems can be solved

There are situations which imply certain restrictions on the factors, which make the problem specification difficult. The high flexibility of STAVEX however allows combining very different requirements. [More...]
You will become more confident in using the methodology if you take part in one of our trainings in statistical Design of Experiments (DoE), where you always are welcome to address your own applications.
We are always at your disposition for any questions. Please call us under +41 61 686 98 77, or use our contact form.

powered by Swisscom AG
c