Cancel
Showing results for 
Search instead for 
Did you mean: 

Interpret results by ExperimentManager

Experimenter
Experimenter
Hi!
I got the results using the ExperimentManager.
I do not have enough understanding of each item.
- Standard deviation, Minimum, Maximum, Left interval bound, Right interval bound
The result is different even if I assign it to 6 Sigma.
Somebody please understand me.

통계 결과값.jpg

4 REPLIES

Re: Interpret results by ExperimentManager

Siemens Pioneer Siemens Pioneer
Siemens Pioneer
Hello! Could you share the model you ran? Without knowing your inputs it is impossible for anyone to interpret the results you have.

Betreff: Interpret results by ExperimentManager

Gears Phenom Gears Phenom
Gears Phenom

@kes  schrieb:
The result is different even if I assign it to 6 Sigma.

And what do you mean by "the result is different"?

___________________________________________________________
Alex Dilg, Consultant at SimPlan AG (www.SimPlan.de)
This post was good and/or helpful to you? Please provide some Kudos, thank you!

Betreff: Interpret results by ExperimentManager

Experimenter
Experimenter
The standard deviation value is 0.2
As shown below, the difference value of each value is not 0.2.
0.1214, 0.1458, 0.1458, 0.1521.
What does the standard deviation 0.2 value mean?
통계 결과값2.jpg
Explain the picture below.
통계 결과값1.jpg

Betreff: Interpret results by ExperimentManager

Siemens Pioneer Siemens Pioneer
Siemens Pioneer

If you are going to use the experiment manager I would strong recommend you to read up about statistics and statistical analysis.

 

And again, without you indata it is really hard for anyone to interpret the results. If you don't feel like sharing your model, at least share the complete experiment report. In the upper left of the report there is a save button, allowing you to save it as a HTML file (pic included).

But I'll try to summarize a bit at least:

 

The standard variation is a measure of how "grouped together" your data points are. If most of the experiments get similar results, then you'll get a smaller standard variation. If the results are spread apart, the standard variation gets bigger.

 

The minimum and the maximum are, as their names implies, the lowest and the highest results you got in all runs.

 

The left and right bounds constrain the confidence interval (compare the values to the confidence interval graph just above). With the interval you have, and the data you put into your model, you can say with 95% confidence that JPH (whatever that is) will fall somewhere between the lower and upper bounds (52.15 - 52.45)

 

Weather these results are accurate or not is also impossible for me to say, as I don't know how many experiments you ran. More runs = more accurate result.