A process is understood before it is controlled. This section climbs the chain: from what the product has to do, to the characteristics that guarantee it, to what makes them vary.
That is the Quality by Design approach, served by in-line measurement. The vocabulary comes from pharma; the reasoning applies wherever a specification exists. PAT-INDUSTRY applies it in pharma, biopharma, chemicals, cosmetics and food.
Five pages, in the order the questions arise
What PAT is
The definition, what the founding text actually says, and three misconceptions.
From product to process
What the product has to do, the properties that guarantee it, the steps that create them.
What impacts quality
Where variability comes from, the IPO matrix, and the trap of a model that correlates without explaining.
Quality by Testing or Quality by Design
Two ways to reach the same conformity, and what the shift changes on the shop floor.
Design space and robustness
How far your process can move before the product stops conforming, and why.
Glossary
Sixty-six terms, written for the process engineer who has to decide, not for the spectroscopist.
Where the technical pages on measurement went
The fundamentals of measurement — mass actually analysed by the sensor, sampling plan, calibration burden, method validation, data integrity — deal with the measurement itself, not with the process. They are written for analytical teams.
They are gathered under Measurement fundamentals and, for everything to do with models, under Chemometrics.
The distinction is not cosmetic. Understanding the measurement and understanding the process are two subjects, and they do not interest the same reader.
Which variation in your process has no explanation?
A drift that keeps returning, a batch that fails without an identified cause, a parameter you suspect without proof. Forty-five minutes are enough to lay out the chain on that one case.