June 16, 2025, 9 - 12 am
Workshop facilitators: Lennart Eriksson, Principal Data Scientist, and Henrik Widmark, Data Scientist at Sartorius Digital Solutions Team
Design of Experiments (DOE), Quality by Design (QBD), and Machine Learning methods, including Multivariate Data Analysis (MVDA), are becoming increasingly important in process development and process monitoring.
DOE is a statistical technique that optimizes resource utilization and extracts valuable information from experimental data. DOE empowers scientists to optimize processes efficiently, leading to better product quality and reduced development costs. DOE is the backbone for efficient QBD implementation strategies. The final specifications for a region in factor space where all specifications on the responses are fulfilled to a defined risk level is called Design Space.
MVDA, an important approach in the Machine Learning ecosystem, is the science of separating the information/signal from the noise in data with many variables and presenting the results in a simple graphical format. Quickly go from a complicated table of numbers to a simple plot of the essentials. MVDA is the key to unlocking the information residing in complex process datasets.
The objective of this workshop is to show how deployment of advanced data analytics aid users in taking informed, data-driven decisions about their analytical systems and manufacturing processes. Although illustrations are mainly drawn from upstream and downstream steps in biopharmaceutical industry, the selected uses cases are illustrative and easy to follow and understand.
The workshop is composed of lectures and demonstrations in the software’S MODDE®, SIMCA® and SIMCA®-online based on real life investigations. The learning material is intended for researchers, scientists and engineers from all sectors of industry and academia. No prior knowledge of statistics is assumed.
Key Learning Points:
- Learn how DOE enables maximally informative experiments
- Understand how to analyze experimental data using sound statistical principles
- Get insights into the fundamentals of QBD and Design Space estimation
- Learn what MVDA is and its relation to Machine Learning
- Find out how MVDA extracts the golden nuggets in your data
- Discover how Machine Learning methods can be used in process development and monitoring
Henrik Widmark
Lennart Eriksson
The Workship is led by Lennart Eriksson, Principal Data Scientist, and Henrik Widmark, Data Scientist, both part of Sartorius Digital Solutions Team. Lennart has worked with DOE and MVDA for over three decades across various industrial segments. His special interest lies in teaching all aspects of DOE and MVDA, ranging from onboarding new practitioners all the way up to the expert level. Henrik has over twenty years of experience from applying DOE and MVDA in product and process development for the biotech industry. Henrik has a special interest in Quality by Design and improvement strategies like Lean and SixSigma.
