Umetrics Suite Blog

Using advanced data analytics to make the shift to continuous process manufacturing

April 2, 2019

Throughout the evolution of manufacturing, many industries have gradually shifted away from batch process to continuous process manufacturing as production technologies matured. This has included industries such as chemical, petroleum, steel, automobile, consumer goods and food manufacturing.

SSingle-use-systems replacing stainless steel in biopharma manufacturing

Stainless steel as a standard facility setup for manufacturing processes in the biopharma industry is losing ground to single-use systems and ushering in a move toward continuous process monitoring using advanced data analytics tools.

With a growing demand for affordable drugs (e.g. biosimilars) and an ever-increasing pressure to reduce manufacturing costs, next generation approaches to manufacturing are now taking off in the biopharmaceutical industry as well. The fast-moving field of advanced data analytics – along with process analytical technologies such as sensors – are contributing to de-risking next generation processes in the biopharma industry.

The evolution of processes and the product life cycle

To better understand why this happens, let's take a look at the typical evolution of production processes in manufacturing.

technology adoption

The typical product life cycle starts with an introduction phase in which a small number of users adopt the new technology. Next, comes a growth phase where more people start using it. Then, it reaches maturation and it becomes a standard that many embrace. There is often a brief period before things start declining. At that moment, industries or technologies need to reinvent themselves and look for innovation – or they face decline.

For example, if we look at stainless steel as a standard facility setup for manufacturing processes in the pharma industry, we can see it’s a bit at the tipping point between decline and innovation. This is mainly driven by the next generation of molecules that need smaller batch sizes aimed at smaller patient populations and personalized medicine or rare diseases (think of bispecific antibodies, antibody fragments, and antibody-drug conjugates).

In addition, we are facing an aging infrastructure of stainless steel plants, and there is a limited capacity within the developing world, which really drives the adoption of new manufacturing strategies.

Migration from stainless-steel to single-use systems

Nearly all commercial manufacturing of biopharmaceuticals to date has involved the use of stainless steel process lines and facilities. Single-use technologies, however, are becoming a more mature market. Fully single use facilities are being built everywhere.

In fact, single-use or disposable bioprocessing equipment is now used for more than 85 percent of pre-commercial pharmaceutical production. Single-use systems (SUS) refers to biopharmaceutical manufacturing (bioprocessing) equipment designed to be used once (or for a single manufacturing campaign) and then discarded.

A couple of reasons for the adoption these new types of facilities include increased flexibility and decreased costs for production, along with growing regulatory acceptance.

For instance, compared to stainless steel, the equipment arrives sterile, which avoids the need for cleaning, sterilization and validation of sterilization prior to usage. This also reduces the need for specialized steam or cleaning systems or having complex plumbing systems installed.

With so much less infrastructure required, the overhead is reduced and the facility footprint can be much smaller.

Single-use has a down side

However, there is a downside to single-use systems: lower output. The maximum operating volume for single use bioreactors is about 2000 liters.

If you look at the classical setup of the single-use facilities where you use six 2000 liter bioreactors feeding into one downstream line, a six-pack of 2000 liter bioreactors would only yield about 20 percent of what a full-scale stainless steel manufacturing plant would yield. This is why the industry is looking at process intensification and continuous process manufacturing to increase the output while reducing the cost of goods for the process.

Two areas that are contributing to the success in continuous process manufacturing, are:

  • Process analytics technologies (sensors)
  • Advanced data analytics

Sensors make it possible to do more inline and online measurements instead of offline measurements completed in a far-away QC lab. Getting the operator’s process information at the point of control makes it possible to design new automation concepts that are driving continuous process manufacturing.

What about the environmental impacts?

Some people might be concerned about the environmental impacts of single use systems by creating waste. However,  utilizing single-use technology can decisively improve the environmental footprint of biomanufacturing processes. Single-use equipment may help lower energy consumption and reduce climate-changing emissions. In particular, single-use technology can dramatically reduce the usage of water, a resource indispensable for life, by more than 80%.¹  

Toward continuous manufacturing

What are some of the reasons that support continuous process manufacturing? These include:

Reduced manufacturing costs — make better use of capital assets installed at your facilities, as well as reduce overall operational expenditures on an annual basis

Improved process control — improve product safety and product quality consistency, an important move in the pharmaceutical industry, which is facing increased public pressure to manufacture affordable drugs for a growing global population, and where safety standards and quality must be the same every single time.

In 2017, the BioPhorum Operations Group (BPOG) (an industry collaboration made up of the most influential leaders of the Biopharmaceutical industry) presented a technology roadmap for the future of biomanufacturing. They described five different manufacturing scenarios in the biopharma technology roadmap. These were:

  1. Large-scale Stainless Steel Fed-Batch
  2. Intermediate-scale Single-Use Perfusion
  3. Intermediate-scale Multi-Product Single-Use Fed Batch
  4. Small-scale < 500L Portable Facilities
  5. Small-scale < 50L for Personalized Medicine

Small-scale facilities can be used in case of global emergencies (such as the Ebola outbreak a few years ago). In addition, there is an increasing interest in small scale production for personalized medicine – where every patient is one batch, which we see for example, with CAR-T processes or autologous cell therapy processes.

Key technology developments that increase the focus on the Intermediate facilities include:

Process Intensification – increase facilities output to lower the cost of goods for processes
Automated Flexible Facilities – which can reduce operation expenditures
Real-time Release – through Inline monitoring, multivariate analysis and predictive modeling

What manufacturing strategies are being used now?

Several major pharmaceutical companies are working on fully continuous processing or hybrid production platforms where, for example, the upstream product is continuously perfused out of the bioreactor while downstream is still done in batches. Several commercial products now on the market use this strategy.

In combination with process intensification, you drastically increase the titer you get from your upstream process, getting more out of a smaller footprint facility.

Another strategy, which more and more companies are looking at thanks to new technologies, is a continuous downstream or connected downstream process, where some of the downstream purification steps are combined into a smaller process.

Eventually, the goal for many manufacturers is to get to a fully integrated, continuous and automated process that makes full use of next generation technologies.

next generation techniques pharma companies

Photo source: Konstantinov, K, Cooney, C, “White Paper on Continuous Bioprocessing”, 2014, Wiley Online Library

Real time monitoring enables continuous manufacturing

To make continuous manufacturing a reality, companies rely on real-time data monitoring and analytical tools that help identify (and even predict) early on when a process is deviating from acceptable parameters.

Multivariate data analytics provides modeling on a number of levels such as:

Modeling a process – which includes data reduction and diagnostics (finding root causes)
Modeling a process output – to monitor the critical quality attributes (CQAs) of the final product
Modeling batch processes –first to monitor the batch evolution and then to monitor specific quality attributes of the final product

One of the values of data analytics for these continuous processes lies in process understanding and control, while also assisting in proper batch evaluations and risk assessment, such as:

  • How quick can I spot possible contamination?
  • Can I plan the number of Protein A or capture chromatography cycles, and associated shift scheduling and buffer preparation, based on predictive modelling?
  • Is the batch meeting the product CQAs?

For example, based on cost models build for an intensified monoclonal antibody process using single-use technologies at 2000L, a 1% increase in upstream titer saves about .20 euros per dose (times over 30,000 doses per batch and 20 batches per year) for a significant savings.


To find out more about how data analytics supports continuous process manufacturing, watch this webinar:

“Using a data driven approach to design next generation manufacturing processes”

Watch the webinar




1. Reference: Sinclair, A.; Leveen, et al.;The Environmental Impact of Disposable Technologies, The Biopharm International Guide, November 2008; Base of the analysis: Typical mAb process at 3 + 2000 L scale

Topics: Manufacturing Processes, Pharmaceutical manufacturing, Continuous Process Manufacturing

Kai Touw

Written by Kai Touw

Kai Touw is a (Bio)Pharma Market Manager at Sartorius Stedim Data Analytics. He is a driven and enthusiastic technology evangelist bridging the world of data science into pharmaceutical and biopharmaceutical processing.

Leave a comment