During biopharmaceutical development large data volumes are generated in many stand-alone devices. Furthermore, with increasing projects and requirements to compress development timelines, high throughput and automation systems are being increasingly implemented. The resulting high pace of data generation augments error rates thereby risking data integrity.
This presentation will provide insights into the strategies and tools for data collection, storage and visualization to transform data into knowledge ensuring leaner, more efficient and integral data generation.
An assessment for each process step and equipment connectivity was performed to map, categorize and highlight gaps during data generation. Based on this preliminary assessment, a data architecture composed of multiple tools was designed to close gaps associated to data traceability, integrity & amp; extractability. Moreover, to support decision-making, multiple modelling tools were developed to focus on specific needs during process development, including process optimisation.
Through increasing instrument and system automation & amp; interconnectivity, data traceability, extractability and searchability are improved, and lab efficiency is increased by >20%. Moreover, with the use of mechanistic models, prediction of culture performance becomes a reality uncovering optimized feeding strategies to significantly increase productivity. In summary, our combined data architecture and process modelling techniques provide better insights thus improving process understanding while accelerating process development.