Transforming data into meaningful information for process and optimisation engineers is simplified with our Data Analysis software.
This suite of tools is focussed around assisting process engineers to address typical process improvement activities, starting with understanding and quantifying measurement uncertainty, baselining current performance, calculating process capability and estimating the potential benefits that could be derived from process and control system improvements
The tools help identify process operating abnormalities, enabling automated events to be developed that can capture deteriorating process performance.
The platform includes links to advanced libraries of open source statistical and Machine Learning algorithms, providing an industrial context for the tools.
Key features include:
Univariate SPC Tools, including Shewhart, EWMA, CUSUM and Western Electric Rules
Univariate signal analysis and Feature detection/extraction
Identification of Bad and “suspect” data, with traceable removal or replacement methods
Filtering, Spike removal, Hampel Filter, Data Masking
Robust environment linking to Python and R libraries for advanced statistical analysis
In built Classification tools; K-Means clustering, PLS-DA, Kernel Density Functions supporting Abnormal Event Analysis
Spectral Data (available December 2019)
This module offers multivariate statistical methods to turn spectral data into actionable knowledge. Well-utilised spectral information will quickly become a valuable asset that enables visualisation of process variations, prediction of product quality and identification of critical process parameters.
Spectra from both offline and in-process instrumentation can be seamlessly cleaned and transformed using chemometric methodology, to develop fully traceable calibration models for real time applications.