.
For a process to be considered "validated", is it a prerequisite to show that the process is stable (statistically in-control)?
[Section 6.3]
“For Process Validation, the process must be shown to be in-control and capable (Cpk of 1.33 or greater).”
[Section 14.1] “The purpose of the performance qualification is to rigorously test the equipment, process and the product being validated/verified and to show that they consistently meet predetermined requirements and specifications while maintaining a state of process control.”
[Section 4.6.1]
“Process capability analysis is the examination of the inherent variability and distribution of a process, in order to estimate its ability to product output that conforms to the range of variation permitted by specifications.”
[Section 4.6.2]
“Process capability analysis is used to assess the ability of a process to produce outputs that consistently conform to specifications, and to estimate the amount of nonconforming product that can be expected.”
[section 4]
“Process and product data should be analyzed to determine what the normal range of variation is for the process output. Knowing what is the normal variation of the output is crucial in determining whether a process is operating in a state of control and is capable of consistently producing the specified output.”
“Process and product data should also be analyzed to identify any variation due to controllable causes. Appropriate measures should be taken to eliminate controllable causes of variation.”
http://www.bcg-usa.com/regulatory/docs/1987/FDA198705A.pdf (superceded by 2011 guidance)
[section VIII B.]
“[i]t may be insufficient to assess the process solely on the basis of lot by lot conformance to specifications if test results are merely expressed in terms of pass/fail. Specific results, on the other hand, can be statistically analyzed and a determination can be made of what variance in data can be expected.”
GHTF/SG3/N99-10:2004 (Edition 2)
http://www.ghtf.org/documents/sg3/sg3_fd_n99-10_edition2.pdf
[Section 3]
“The product should be designed robustly enough to withstand variations in the manufacturing process and the manufacturing process should be capable and stable to assure continued safe products that perform adequately.”
[Section 5.5]
“PQ considerations include … assurance of process capability as established in OQ; process repeatability, long term process stability”.
“Process and product data should be analyzed to determine what the normal range of variation is for the process output. Knowing the normal variation of the output is crucial in determining whether a process is operating in a state of control and is capable of consistently producing the specified output.”
“Appropriate measures should be taken to eliminate controllable causes of variation.”
[Annex A – section A.2]
“Reducing variation requires the achievement of stable and capable processes.”
Section 1.2 Exploratory Data Analysis (EDA) Assumptions
http://www.itl.nist.gov/div898/handbook/toolaids/pff/ehb-chapters-1-8.pdf
1.2.1 Underlying Assumptions
There are four assumptions that typically underlie all measurement processes; namely, that the data from the process at hand "behave like":
1.2.2. Importance [of underlying assumptions]
Predictability and Statistical Control
Predictability is an all-important goal in science and engineering. If the four underlying assumptions hold, then we have achieved probabilistic predictability--the ability to make probability statements not only about the process in the past, but also about the process in the future.
In short, such processes are said to be "in statistical control".
Validity of Engineering Conclusions
Moreover, if the four assumptions are valid, then the process is amenable to the generation of valid scientific and engineering conclusions. If the four assumptions are not valid, then the process is drifting (with respect to location, variation, or distribution), unpredictable, and out of control. A simple characterization of such processes by a location estimate, a variation estimate, or a distribution "estimate" inevitably leads to engineering conclusions that are not valid, are not supportable (scientifically or legally), and which are not repeatable in the laboratory.
Also see:
Section 4.2. Underlying Assumptions for Process Modeling
http://www.itl.nist.gov/div898/handbook/toolaids/pff/ehb-chapters-1-8.pdf