Collecting Data (how often?)
- June 3, 2010 at 6:43 pm #53475
Hello everyone! I have searched about this issue, but hasn’t found anything yet. Maybe you can help me…
there is a process that it’s already collecting data, but it’s collecting data every 1 second (let say Fluid Temperature). It’s possible to change the frequency of data collection, from every second to every hour or day (measure the Temperature at the time specified).
How often should it be measured and collected? Is there any tool or test to see what is the optimum collection frequency, based on how the variable change and the significance of the change? if so, How to use it and how to understand it and interpret it?
Roger.0June 9, 2010 at 9:04 pm #190315
Robert ButlerParticipant@rbutler Include @rbutler in your post and this person will
be notified via email.
If you have reason to believe that the data you have collected is representative of the process then the easiest thing to do would be to build a program to run resamples of the existing data for a variety of time intervals (i.e. every tenth data point in time sequence, every 100th point, etc.) and then run the appropriate tests on these subsamples and see how the results compare with the analysis with all of the data. When you reach a point where the differences in results matter you could use that subsample frequency as a starting point for further investigations into the possibility for reduced frequency of data collection.0July 1, 2010 at 4:38 pm #190392
You must be logged in to reply to this topic.