New technologies in process automation are the basis for end-to-end data streams

If companies in the process industry - and the chemical industry in particular - want to produce in cycles in the future, there is no way around end-to-end networking. The foundations for this have already been laid for several years: Digital twin, administration shells, Namur Open Architecture or Ethernet APL are the buzzwords that are electrifying process automation engineers. However, the technologies under discussion are so diverse that some observers have lost sight of the big picture. But not least at ACHEMA 2024, it became clear that many things that were previously ridiculed as the playground of tekkies now have their place under the umbrella of digitalisation.

So let's start with the big one: the digital twin. For more than a decade, automation engineers and digitalisers have been expecting great benefits from mapping real systems in the digital world: it can be used to simulate scenarios and options for process optimisation, prospective system operators use it for training purposes and maintenance engineers supplement the twin with analysis tools in order to identify maintenance requirements at an early stage and optimally plan maintenance.

It is important to define the term "digital twin" clearly - because it is ambiguous. It is often used to refer only to the simulation model, which is created and offered in separate formats. Digital twins based on the asset administration shell differ significantly here. Administration shells are being significantly further developed by the Industrial Digital Twin Association (IDTA). The aim of the IDTA is to establish the asset administration shell and the further development of submodels. Administration shells as a data model are characterised in particular by their interoperability. By using classifications, e.g. ECLASS, all information can be easily imported into software environments such as an ERP or asset management system.

Even if process industries such as the chemical industry have not yet succeeded in using the digital twin across the board, the topic is now making progress. More and more managers in the processing industry are realising that end-to-end data streams make many tasks easier. Especially those that arise with the transformation away from the linear economy (take-produce-dispose) and towards a circular economy. In the circular economy, products at the end of their life cycle become raw materials for new chemicals. To ensure that the necessary data is available at all times, the digital twin will have to virtually map the entire supply chain in future.

The topic is now receiving an additional boost from the EU Commission's requirements: the new ESPR Directive (Eco-design for Sustainable Products Regulation), which is due to come into force in 2026, requires a digital product passport that contains comprehensive information about the lifespan of products and their ecological footprint. Without this, products may no longer be placed on the EU market from 2026. Although the directive does not specify in detail how this digital product passport should be realised, the aforementioned solution consisting of a digital twin and administration shell is the perfect solution.

In a circular economy, stakeholders across company boundaries are interested in where which feedstocks are available and in what quality. Disruptions in the supply chain must be taken into account as well as their consequences for production and supply capability. If this data is transparent and available in real time, production processes can also be planned, retooled or adapted to changes in raw materials. Transparency is also important so that emissions can be precisely balanced for each product - this not only prevents greenwashing that distorts competition, but also makes it easier to draw up sustainability balances - on the basis of which the right decisions can be made to achieve the net-zero target.

In addition, there is another level of complexity: the transformation of the energy system is leading to more and more processes being converted from fossil fuels to electricity from renewable energy sources - the electrification of the chemical industry is already in full swing. However, electricity from wind and solar energy is not continuously available. Where plants were previously planned for continuous operation on the basis of energy sources and raw materials that were available at all times, in future operators who flexibly align their processes with the energy and raw materials available will be able to realise economic benefits. Added to this is the overarching goal of linking and jointly optimising the energy sector, industry, transport and buildings in the sense of sector coupling. This dimension also makes the management of chemical production more complex. And here, at the latest, it becomes clear that humans alone are overwhelmed by the abundance of optimisation goals and influencing variables: the key to holistic optimisation therefore also lies in digital tools.

The industry is expecting great benefits and support in decision-making from new tools based on machine learning and artificial intelligence. These play to their strengths wherever large amounts of data are available - and this is the crux of the matter: although chemical companies are already producing more data than ever before, the data is often inconsistent and frequently lacks context - for example, about the relationships in material cycles. The problem is well known and the industry has long been working on solutions: In the DEXPI initiative founded by BASF, Bayer and Evonik in 2011, for example, a neutral data format was defined with which process information can be exchanged between software products from different manufacturers of engineering tools. The current Manufacturing X initiative goes one step further: the stakeholders of the Industry 4.0 platform have set themselves the goal of realising the Industry 4.0 data space and the transformation to a digitally networked industry across the board. Ultimately, the aim is to create a data-based economy.

However, the same basic prerequisite applies to all of these initiatives as to the circular economy: they require transparency about processes and resources used throughout the entire value chain; and they also require data to be shared across company boundaries without this leading to a loss of expertise within the companies. And the benefits of digitalisation - for example in production - often arise in a completely different area; without a holistic view, process operators often lack the business case and therefore the motivation to invest in digitalisation. There are even data gaps within companies: a now classic example is the information technology gap between plant planning and the operating phase: in digitalisation projects, it is often necessary to painstakingly reconstruct data that was already digitally available in the planning phase but was not passed on in a digitally readable format due to different responsibilities. In a future circular economy, information about a product must also be passed on throughout its life cycle - here, the use of blockchain solutions could ensure consistent information flows in the value chain. The examples show: Those responsible for digitalisation must focus on the entire value chain.

In this context, the role of process automation is also changing: on the one hand, they create the conditions for consistent data flows, while on the other hand they help to manage the data flows. The blurring of the boundaries between IT (information technology) and OT (operational technology) can already be observed in many chemical companies - Industry 4.0 and the digital transformation are leading to an increased convergence of IT and OT and the integration and networking of systems. In the chemical industry, however, this must always take place against the backdrop of high security requirements.

The current initiatives of process automation companies are also part of the target image of a digitalised chemical industry: Ethernet-based communication technology Ethernet-APL (Advanced Physical Layer) is intended to help the digitalisation of the field (sensors, actuators) achieve a breakthrough. Numerous solutions - both in the form of field devices and data transmission - were already on display at ACHEMA 2024. The mechanisms for this to work across manufacturers have now been established. The Profinet International (PI) organisation is responsible for this, and another two projects were recently placed in its hands: Namur Open Architecture (NOA). This enables data that was once stranded in field devices to be used for optimisation applications - for example in cloud applications - for example to detect anomalies or to reduce maintenance costs, energy and raw material consumption.

The second project promises flexibility, the module automation approach: system modules are described in a standardised way via the Module Type Package (MTP) - this massively reduces the integration effort when assembling modular system components. MTP is expected to save up to 70 % of engineering time alone. And another acronym is important in this context: OPA-S. The Open Process Automation Standard is currently being used to create a standardised and manufacturer-independent architecture for the automation of process plants. The main aim is to promote interoperability between devices and systems from different manufacturers. This makes it easier for plant operators to scale and expand their systems and access data from different sources.

Continuous digital data streams are an important prerequisite on the way to a climate-neutral economy. Process automation and the digital twin are central building blocks of digitalisation, and their implementation must go far beyond pure plant automation in order to map the entire value chain. Current process automation initiatives and technologies such as Ethernet APL, NOA, MTP and OPA-S help to provide data from the production processes.