Process Parameter Analysis and process understanding — Some industrial examples

At the end of process development or during a routine production period often rises the question to prove the identity or the significance of difference of processes or batches. To answer these questions objectively, relevant process data collected by computerised production equipment with recipe control and statistical tools are essential.Attempts have been made to define some quantifiers to make process parameters comparable and to characterise processes. Three types of processes and equipments have been discussed and compared on the level of Accuracy, In-Process Precision and Repeatability of Input and Output parameters. The processes are: Coating Process 1: Driam Vario 500/600 perforated drum coater and Fluid granulation Processes 1 and 2: Glatt WSG 15 top-spray granulator (version 1 with blow-off metal filter and version 2 with one chamber textile filter and mechanical shaking).Computer programs have been developed to structure, organise and statistically evaluate process raw data collected by the above mentioned equipments. The results are presented in uniform summary tables for every batch following the recipe structure phase by phase.The programs perform further statistical evaluation of summary tables, producing control charts.These methods and tools are useful to filter out differences and unusual behaviour, to track back the original process, search the reasons and to understand the process.The introduced Process Parameter Deep Analysis give the possibility to investigate filmcoating (or fluid granulation) process through the wetting curve (average moisture content vs. time).The method helps process understanding and design of process optimisation.

Computer programs calculating Accuracy, In-process Precision and Repeatability were developed to perform statistical analysis on process data collected by computerised equipment, e.g. fluid granulator and film-coater with recipe control. The programs perform further statistical evaluation, producing control charts, to filter out unusual behaviour and track back the original process for better understanding.Figure optionsDownload as PowerPoint slide