Theory of Sampling (TOS), Process Analytical Technology (PAT) and Chemometrics
Industry is increasingly recognising the importance of detailed process information provided by appropriate sample extraction and/or sensor signal acquisition along the full value chain, enabling process automation and informed decision making at all organisational levels to get the most out of mining or processing assets. With expertise in proper sampling, material characterisation, sensor technology and data analytics, the SIX-S team is the right partner to provide you with the reliable information you need to make informed technical and business decisions.
Process Analytical Technology (PAT) involves chemometric analysis and modeling based on appropriate sensor data. When properly applied, PAT can complement and enhance the process knowledge associated with traditional sampling and laboratory testing. Both sample and signal data must be obtained through representative sampling that ensures accuracy in terms of materials composition and lot characteristics, and precision in terms of minimising sampling, handling and analytical errors. Chemometric data models must make use of mandatory performance validation, e.g. in terms of prediction and classification (unrestricted test set validation).
Without representative process samples, the chain of evidence relating to lot characteristics is poor due to inherent material heterogeneity at all scales. This applies to both samples and sensor signals. It is not possible to correct deficiencies in representativeness in the subsequent analytical modelling of the data.
- Audits and design assistance
- Sensor/ Process Analytical Technology calibration and validation.
- Test work design and management
- Establishing sampling quality assessment, assurance and control protocols (QA, QC,QM).
- Training, courses, mentoring