Internet of Things: A New Era for Biomanufacturing?
Duygu Dikicioglu and Lidia Borgosz explore the opportunities and challenges that will determine the future of the Industrial Internet of Things (IIoT) in bioprocessing
INDUSTRY 4.0 opens a new dimension of potential improvement in productivity, flexibility, and control, with the end goal of creating smart manufacturing plants with a web of interconnected devices. However, the nature of biomanufacturing amplifies the complexity and risks associated with the adoption of Industry 4.0 technologies, including IIoT, compared to other industries.
The unique nature of the industry, which uses living systems or their components in manufacturing, necessitates tight regulatory controls around production. These regulations dictate how IIoT implementations can be introduced to bioprocess industries. While some of these considerations are not unique to bioprocessing, in many cases they are notably different to those concerning chemical process industries.
These constraints are not only relevant concerns for Industry 4.0, but they have also restricted the adoption of Industry 3.0 technologies, including automation and robotics.
However, progress is being made, and the universally touted benefits of adopting Industry 4.0 solutions are increasingly being realised. Industry 4.0 solutions, such as process analytical technology, design-of-experiments, and multivariate data analysis are already being implemented to help resolve challenges around ease of connection, real-time monitoring, and data collection. They are also assisting in increasing efficiency and minimising uncertainty stemming from human intervention, which is among the key considerations for the implementation of IIoT. Although these methods have been employed in bioprocessing for a number of years, they do not yet attempt to modify the process as a whole, but rather assist the implementation of IIoT. The further adoption of IIoT can therefore help bioprocessing industries create a holistic view from past, current, and future perspectives, through analytics and predictions. This helps the industry to be proactive, rather than reactive, with the goal of improving any bioprocess operation.
Some of the Industry 4.0 concepts, including soft sensors, hybrid modelling, digital twins and machine learning that are directly linked with the concept of IIoT are becoming increasingly prominent in the bioprocess industries. The capabilities of smart devices within the manufacturing process, such as soft sensors, are not limited to measuring process parameters but can also be used for tracking and tracing the data collected, detecting abnormalities, and assessing instrument availability in real-time, especially when combined with analytics, edge and cloud computing, as well as machine learning and AI.
However, these technologies must be applied with caution. Some bioprocess parameters that are difficult to measure render scientists reliant on predictions, which require large sets of data for accuracy. It should also be noted that each step of a bioprocessing operation, regardless of employing IIoT components, requires a dependable, well-defined, and well-characterised control mechanism. This must allow critical process parameters to be monitored (while ensuring that product specifications are met), detect deviations, and optimise the process. This strict restriction stems from the tight regulatory requirements around a substantial number of operations that are active in the healthcare domain.
Glossary
In the simplest terms, Internet of Things (IoT), is a network of interconnected “things”. The terminology dates back to 1999 but the concept dates back even further; its true history began with the internet itself and the concept of exchanging information from machine to machine without human input. Today, IoT describes a collection of technologies that, when combined, produces a system of interconnected devices and entities generating information that is more powerful than the sum of its parts. Industrial Internet of Things (IIoT) is the utilisation of IoT technology in an industrial setting.
Biomanufacturing
Using organisms, such as bacteria, fungi, algae, mammalian or plant cells, or their components to manufacture a variety of different products, from pharmaceuticals, vaccines and foods to commodity and high value chemicals and biofuels.
Hybrid modelling
A modelling approach where first principles (mechanistic) models are complemented with data-driven models. The hybrid approach provides an alternative when there is a lack of understanding of mechanistic details, and at the same time offers a more robust alternative to inferred data.
Digital twins employed in manufacturing are virtual representations of production systems that mirror the behaviour of physical system parts and their relationships.
Machine learning is an artificial intelligence field that focuses on the development and study of statistical algorithms, which can learn from data and generalise to unseen data, and thus perform tasks without explicit instructions.
Edge computing is an emerging computing model in which the networks and the devices are near the user, which allows data processing to take place close to the point where it is generated and collected. This leads to improvements in processing speed and volume.
Cloud computing
The delivery of computing services including but not limited to data storage and processing power through the “cloud”; a global network of remote servers.
Industry 4.0 solutions, such as process analytical technology, design-of-experiments, and multivariate data analysis are already being implemented to help resolve challenges around ease of connection, real-time monitoring, and data collection
Role of hybrid process modelling and digital twins in IIoT
Recent Editions
Catch up on the latest news, views and jobs from The Chemical Engineer. Below are the four latest issues. View a wider selection of the archive from within the Magazine section of this site.
