Master your processes by analyzing them!
For an understanding and supervision of processes
At a time of digitalization in industries, it is essential to better understand production processes (Process Understanding), in order to identify quality leverages of the products being manufactured, the key parameters (Critical Process Parameters – CPPs) of the production, from raw materials to finished products.
This increased knowledge of processes then makes it possible to optimize manufacturing process by supervising the processes through the measurement of these CPPs in the context of Industry 4.0. This control then makes it possible to increase productivity, minimize the risks of production outside specifications, and reduce over-quality while minimizing material losses and the environmental impact.
Vous souhaitez :
Understand
your process using Data Mining methods
(Process Understanding)
Supervise
your processes for detecting drifts in real time (Process monitoring)
Control
your process by coupling your models to the automatic control loops (Process Control)
For many years, Statistical Process Control (SPC) has made it possible to supervise and control processes in real time thanks to Control Charts for each CPP, in many industrial sectors (agro-food, chemistry, petrochemistry, cosmetics, …).
In Industry 4.0., more advanced process monitoring techniques are now applicable, in order to integrate all the data available on a process (Industrial IoT, online analyzers, process parameters).
The fusion of all these multivariate sensors and signals then generates new advanced control charts, making it possible to analyze and control the processes by synthesizing all the CPPs at a glance, whether they are batch (BSPC) or continuous processes (MSPC).
Within the framework of multivariate modeling techniques for process supervision, the methods applicable to continuous processes must be separated from those dedicated to batch processes.
- For continuous process supervision, MSPC – Multivariate Statistical Process Control – method was developed to set up multivariate control charts, statistically delimited by confidence intervals.
- For batch process supervision, challenges in terms of data analysis are more complex because of the additional dimension of time, and the evolution of the state of the process during this time (batch kinetics). BSPC – Batch Statistical Process Control – method is used.
Software implementation
We develop your calibrations with all the Chemometrics software on the market (Unscrambler®, SOLO® or PLS_Toolbox® in the MATLAB® environment, SIMCA®, etc.), as well as those provided by spectroscopy equipment manufacturers (OPUS, WinISI, NIRCal, TQ Analyst, etc…), but also free software (R, Python…).
Our expertise for the analysis of your process data
With over 15 years of experience in data analysis (Chemometrics and Machine Learning), in particular applied to at-line, on-line and in-line measurements in processes involving online analyzers, process data and sensors IoT, the multidisciplinary experts of our teams accompany you at each stage of your process data analysis projects.
They talk about Ondalys
“Ondalys, to validate the way we analyze data”
