Hi, I've become fairly confused over the notion of "offline (engineering) analysis" :confused:

(Disclaimer: I don't have much background in Quality Management in manufacturing, as demonstrated by the fact that I'm not even sure if the terms "offline analysis" and "engineering analysis" are interchangeable.)

I've tried to look online (including on this site) for a description of what this type of analysis involves, when and on what types of data it is performed, and its main benefits, but cannot seem to find a resource that could summarise what offline analysis entails. I can mainly find information on the Taguchi method, but surely this is not all there is to it?? I've also found references to various manufacturers' software packages that do both "online/real time and offline analysis", but they have been no help either.

I'm especially looking for examples of offline analysis in medical device/diagnostics companies, but any information would be much appreciated!



My apologies if I have posted this in the wrong forum.

If you're running a continuous process from which you're collecting data in real time, you can do "real time analysis" of that data in nearly-real-time to gain an understanding of what the process is doing, but such analysis must be done pretty quickly in order to remain synchronized with the data's arrival. Or, you can do "offline analysis" of a selected amount of data, which necessarily represents a time before NOW but which can be analyzed in more depth.

Real time analysis may be used for monitoring processes that can change very rapidly, and for which past data is of limited or no importance as an indicator of the process's present or future status. Offline analysis may be used for monitoring processes that change slowly, and for which study of past events and trends can reveal important information.

Thanks MIREGMGR!

One question regarding the following in your explanation:
"Real time analysis may be used for monitoring processes that can change very rapidly, and for which past data is of limited or no importance as an indicator of the process's present or future status. Offline analysis may be used for monitoring processes that change slowly, and for which study of past events and trends can reveal important information."
(I don't know how to properly quote another post so I just copy-pasted)

But can't one use data to do both online and offline analysis, for example:

- Online analysis: data from a particular production line can be analysed online in real-time and used by the production department to be able to immediately react if the process becomes out of control or is headed that way

- Offline: this same data gathered over a longer period of time can be analysed offline, perhaps in combination with similar data from other production lines, and used by e.g. the R&D department for product development purposes?

Did I get that right?

Yes, agreed.

Data can be used in any way that it is relevant and valid.

To recap, I think in your context, "offline" means "analyzing data that was already collected, as opposed to data that's being analyzed almost simultaneously with being collected", i.e. in near-real-time.