The two objectives of the bioprocess project were data flux from sensors and devices into a database for subsequent analysis and process data analysis using a central data base. The major aim of the first objective was the establishment of a seamless data flux from devices and sensors into a central data base. To this end, we have first built a data model describing the connection and interrelation between files generated at different time points and on different devices. Next, we generated reference data to identify where and how metadata for automated file parsing should be placed and structured. Finally, we have built a tool that automatically processes and converts the different output files, file types and formats from the different devices into a uniform meta-structure. The aim of the second objective was to assess the data derived from our model process with respect to statistically significant differences. The goal was to do so in an explicit manner, i.e. not to use artificial neural networks, and to assess the quality of the according process models. Next, we systematically analyzed which types of verification data are best suited to test the model performance and finally to allow for an automated data processing. Importantly, the above steps focused on chromatographic separation, i.e. only a small fraction of the process data, which was due to overall process complexity and will require to reiterate and generalize the steps for other types of data.

Get in touch

Contact us to get in touch! With a membership, you’ll gain full access to all project information and updates.