What is the difference between OOS/OOT

invitro_spain

Quite Involved in Discussions
#1
The terms OOS and OOT always emerge in connection with the handling of deviating analysis results during Quality control processes. Both terms have the following definitions:


Out-of-Specification (OOS) Results
A result that falls outside established acceptance criteria which have been established in the Company SOPs.


Out of Trend (OOT) Results
A time dependent result which falls outside a prediction interval or fails a statistical process control criterion.
A trend is a sequence of temporal procedures, e.g. for the manufacture of different batches of a product. There are two types of trends:
 

cal guy

Starting to get Involved
#2
Example of OOS would be; a single calibration is performed on a piece of equipment. The readings of the UUT and Cal Standard differ by more than the allowed accuracy tolerance per Company SOP. The equipment that was calibrated is not meeting the "established acceptance criteria" and is therefore, Out of Specification (OOS).

Example of OOT As it states in the description, OOT is time dependent. As a cal guy (technician), the way I would use trending would be to track calibration results. If you are trending previous years of calibration results for the piece of equipment in the above example, the gauge may have shown a trend over the years of the error increasing. This is an indicator to replace the gauge before an OOS finding. This may not be the example you are looking but it's how I trend calibration data in the cal lab.
 
#3
OOS is the comparison of one result versus specification criteria, while OOT is the comparison of many historical data values versus time.

For an example consider a limit for the test is between 94.0-110.0%

For the above test, the result obtained is 93.2,which is out of the specification limit(OOS).

For the same case, the result obtained is 95.7, which is within the limit. Although the result is within the specification, we should compare the previous batches and we have to calculate the average value which is called out of trend(OOT).
 
Top