Demonstrate how machine learning tools can aid development of formulation, analysis and manufacturing of pharmaceutical drug products

Novo Nordisk has initiated several optimisation projects using machine learning tools and robotics as alternatives to the traditional design of experiments (DoE) methodologies and hands-on experiments. The reasoning is a pursuit to find better optima using fewer resources. In this webinar, Rolf Hejle Taaning and Erik Skibsted from Novo Nordisk will demonstrate optimisation of pharmaceutical liquid formulations and chemical reactions using a mix of different tools like Bayesian Optimisation, Principal Component Analysis, and Robotics.
Presented by Erik Skibsted, Principal Scientist Ph.D, Novo Nordisk Denmark

Erik Skibsted received a master degree in chemical engineering from the Technical University of Denmark in 1998 and worked with development of novel fluorescence sensors for bioreactor monitoring, before starting at Novo Nordisk in 2000. In 2005 he finished his Ph.D. studies at Novo Nordisk and the University of Amsterdam in the group of Process Analysis and Chemometrics under the guidance of Professor Age Smilde. In his Ph.D. work, he developed novel algorithms for spectroscopic data analysis and published several papers on the topics of process analytical technology applied to pharmaceutical drug product manufacturing.
He has now worked almost 20 years at Novo Nordisk with spectroscopy, chemometrics, analytical method development, troubleshooting of biopharmaceutical manufacturing processes, powder and tablet characterization, and lately implementation and validation of near-infrared spectroscopy in continuous blending processes and machine learning for optimization projects.
He is deeply interested in the integration of machine learning, AI, chemometrics, sensor technology, and computer technology into drug product manufacturing and formulation research.
and Rolf Taaning, Development Specialist Ph.D., Novo Nordisk Denmark.

Rolf Taaning received his Ph.D. in chemistry from the University of Aarhus, Denmark in 2010 for work within the field of organic transformations with single electron reducing agents, followed by a three year post doc in the group of professor Troels Skrydstrup working with the development of palladium catalysed carbonylation chemistry and isotope labeling. During this period he co-founded, SyTracks, which provides laboratory equipment and reagents for the safe use of dangerous gaseous reagents in the synthesis laboratory under the tradename ®COware.  In 2013, he joined Novo Nordisk and is today working as a Development Specialist in the department for Chemical API Development.
He has now worked 7 years at Novo Nordisk with chemical process development, scale-up of fully synthetic and recombinant processes, establishing processes for the market at CMO’s, reaction kinetics, Design of Experiments, optimization algorithms.
His current work, beyond his daily obligations to process development, focuses on establishing an automated and autonomous platform for chemical optimization and facilitating the access to optimization algorithms to non-programmers through tool development and graphical user interfaces