Manufacturing in the Industry 4.0 era: how to balance innovation with performance while not being disrupted
10th March 2021 | 10:00am EST | Christos Varsakelis, Senior Manager Global Analytics and Innovation, GlaxoSmithKline Biologicals SA, Mark Demesmaeker, Head of Data Analytics at Sartorius Stedim Biotech and Roger-Marc Nicoud, President & CEO Ypso-Facto |WATCH FOR FREE
The manufacturing status quo reflects a compromise between innovation and performance. But the battle between these values is perpetual and, in the dawn of the Industry 4.0 era, a new compromise between technological progress and the need for stationarity is being crafted. Being able to successfully harvest the benefits of the fourth industrial revolution requires a) separating facts from fiction and b) getting acquainted with the never-before-encountered bottlenecks that will have to be deal with. In this talk, we will go through the unique opportunities and the unprecedented challenges accompanying the fourth industrial revolution. The point of view, however, is not that of the salesman, which is frequently encountered online, but rather of the buyer, i.e. manufacturing. The objective, therefore, is simple, albeit arduous: transform the buyer of this negotiation table into an informed investor that no longer faces a competitive disadvantage.
Presented by Christos Varsakelis, Senior Manager Global Analytics and Innovation, GlaxoSmithKline Biologicals SA
Christos Varsakelis leads the AI/ML compound profiling team in the Janssen Pharmaceutical Companies aiming at the acceleration of drug discovery & development timelines. Prior to joining Janssen, Christos was a Senior Manager of Global Analytics & Innovation in GSK Vaccines focusing on in silico process development. Besides working in the pharmaceutical industry, Christos has an accumulated professional experience that spans across Geoconsulting, the stock market, and the chemical industry. Christos is in the hold of a Ph.D. in fluid mechanics at UCLouvain in Belgium. He is also in the hold of a B.A. in Mathematics from the Aristotle University of Thessaloniki, Greece, and an M.Sc. in Applied Mathematics from the ETH Zurich, Switzerland. His motto is innovation = cash.
Followed by Mark Demesmaeker, Head of Data Analytics at Sartorius Data Analytics (SDA).
Dr. Mark Demesmaeker is Head of Data Analytics, Sartorius Data Analytics (SDA), part of Sartorius-Stedim Biotech. Mark joined SDA in 2018. He is a life science professional with over 20 years of experience in pharmaceutical R&D, analytics and business intelligence, and has held leadership positions at the World Health Organization, Spotfire, TIBCO Software, IBM, and Integrated Clinical Systems. Most recently, he served as Vice President, Clinical & Translational Analytics at PerkinElmer Informatics. Mark holds a Ph.D. in drug metabolism and pharmacokinetics from the German University of Kiel.
In-silico process development and smart manufacturing have already established themselves as competitive differentiators before COVID-19. Now with the pandemic, Biopharma is even more forced to rapidly respond to shifting market needs through the development of novel vaccines and therapies at a scale and velocity that has been unseen before. Some of our partners have managed to respond quickly to this challenge and are now emerging from the crisis as strengthened and well positioned to tackle future challenges. Process simulation in process development, as well as advanced process control in commercial manufacturing, have proven as enablers for decreased time-to-market as well as improved yield and quality.
As expression systems are being pushed to their limits in modern intensified processes, there is an increased need for obtaining improved observability and predictability of their metabolic activity and performance. Hybrid models can provide the necessary input for adjusting feeds and process conditions. In Downstream processing, the adoption of model predictive control mechanisms and tools is also rapidly gaining momentum. Our customers are applying multivariate statistical process modeling in DSP steps, such as protein A chromatography, Ultrafiltration/Diafiltration, or Tangential Flow Filtration, just to name a few examples. Kinetic models based on resin- or membrane characteristics are also being introduced in order to ensure optimal use of expensive chromatography assets.
Using digital tools to speed up process development: a few simple and hopefully rational considerations.
Presented by Roger-Marc Nicoud, President & CEO Ypso-Facto.
Roger-Marc Nicoud is the CEO and founder of Ypso-Facto, a service company offering consulting assistance and software solutions to industrial firms to develop, optimize and secure their chemical and bioprocesses. With the vision to develop comprehensive solutions to produce bio- and synthetic molecules, he founded Novasep in 1995, which became a leader serving the life sciences industry with a recognized portfolio of innovative technologies. He held the position of CEO of Novasep until October 2012, and Chairman of the Novasep Supervisory Board until February 2014. He holds a Ph.D. from the University of Lorraine in process simulation for the nuclear industry. He has published the book Chromatographic processes: modeling, simulation, and design in 2015.
“Manufacturing in the industry 4.0 era: how to balance innovation with performance while not being disrupted?” is thought provocative. How could a great digital innovation associated with industry 4.0 be disrupting and not simply increase performance?
The relatively low adoption of digital tools in the biotech industries should push us considering that current tools may not fully match the real needs. The low success rate during clinical trials requires the fast development of many processes for (mostly) unsuccessful drug candidates: Biotechs need to be agile to save time in a low visibility environment.
In this context, modeling tools can help minimize the experimental burden required for process development by limiting the number of experiments to be carried out.
We will distinguish between two main approaches for modeling:
- Statistical approach: different datasets are represented to the best possible extent by mathematical expressions regardless of the underlying Physico-chemical phenomena.
- Mechanistic approach: the determination of a limited number of parameters associated with a model based on first principles allows to predict of many experiments, possibly outside the investigated region.
How to choose one approach (or both, or none!)? Definitely, the actual need (scaling-up, optimizing, etc.), the knowledge (a basic stoichiometry, separation principles, etc.), and the data available (only production results, a few well thought experiments, etc.) should determine the answer.
By using a few simple illustrations, we will highlight the merits and limitations of these approaches, and their impact on experimental burden savings.
Sponsored by Sartorius Stedim Biotech & Ypso-Facto
Sartorius Stedim Data Analytics develops software under the Umetrics Suite for the design of experiments and multivariate data analysis, for the individual user as well as for online continuous and batch processes.
The stability of Umetrics founding principles and the flexibility of our services enable us to promote every way of implementation of design of experiments and multivariate technology that our clients require. We are committed to supporting our clients in their mission to take control over their data flow, by conveying our advanced expertise in advanced data analytics technology. Our mission is to provide comprehensive solutions which create value from data, enabling better decision making and process excellence.
Our market exists in every sector of industry where there are manufacturing, testing or data analysis activities, but historically the highest number of collaborations has been in the life science, semiconductor and chemical industries.
Umetrics is since April 2017 owned by Sartorius Stedim Biotech AS.
Ypso-Facto is a service company offering scientific and technical consulting, and simulation software for your (bio)chemical processes.
We combine strong industrial experience and scientific expertise with complementary know-how in chemistry/biochemistry, software development, and process engineering to develop disruptive simulation software and support innovative processes.
Ypso-Facto offers a unique GPX® approach combining your expertise (Guess), Predictive mechanistic modeling (P), and eXperimental data (X) to address your challenges.
Simulation Software :
We believe that process modeling and simulation is key to design and optimize robust (bio)chemical processes, and move away from the traditional trial-and-error approach.
Hence we designed some innovative software solutions:
- Ypso-Proxima ®, a Process Evaluation software suite, from experimental Data acquisition, through Process Design, to Cost evaluation.
- Ypso-Ionic ®, for ion exchange processes simulation.
- ChromWorks ™, for chromatographic processes simulation.
With this powerful and user-friendly software, Ypso-Facto aims at helping industrial firms enter a new era for the development and production of their (bio)chemical products.
We also propose a custom software solution for modeling proprietary or non-conventional processes.
Consulting expertise :
We can assist you through all phases of your project development, from brainstorming to scale-up, debottlenecking, and troubleshooting, but also economic evaluation. www.ypso-facto.com