Data mining for bioprocesses

Automated data mining, quality control and supervision

Due to more analytical equipment being fitted in plants and new analytic devices being developed, more and more measurements are available for use. In order to be able to convert this data quickly into knowledge and starting points for process improvements, we have developed Module M1 of the BioAssistent. It allows an easy preparation, administration and evaluation of process data. This way data from different sources can be linked, analysed and compared instantly in relation to the batch.

Task Description
Data preparation
  • Measurements from different data sources can be imported and filed in relation to the batch.
  • In the course of this process they are being prepared by means of various methods to ensure better analysis.
Data mining
  • The major advantage of BioAssistent is that measurements are automatically linked with each other during the data mining process. Additional process parameters and important variables can so be determined more easily. The result is direct access to variables with physical and physiological meaning (i.e. stoichiometry and kinetics).
  • This way the informational content of the data and its validity increase. Processes become more transparent and a greater insight is achieved. Additional knowledge can be gained, and more starting points for improvements be found.
Presentation of results
  • All data can be presented graphically.
  • In particular a comparison of data of different batches is possible within one graphic. This way discrepancies and trends can be identified at a glance.
Online Module
  • The online link is available as an optional extension. All functions are then applied automatically as soon as a new batch or experiment is carried out.
  • All variables are being examined for discrepancies from a normal process run and operating staff are being informed if any anomalies are detected.