Making the Invisible Visible

Guidelines like the European Green Deal, the Ecodesign for Sustainable Products Regulation (ESPR), the Digital Product Passport (DPP), the Corporate Sustainability Reporting Directive (CSRD), and the EU’s energy efficiency directives are making reporting requirements for manufacturing companies in all industries tighter. They are also expanding the focus to the entire life cycle and deep into the supply chain. At the same time, cost and supply risks are increasing as resources remain scarce and the energy market remains volatile.

Companies must continue to face the challenge of using energy and resources more efficiently, not only to reduce current costs, but also to mitigate future price fluctuations. Most companies have already taken low-threshold measures to reduce energy and resource consumption. The lighting has been upgraded, and the cooling systems have been optimized. What’s next?

An automotive supplier has hundreds of metal sheets roll off the production line every day: They are first heated in a furnace, then hot-formed in a press, cooled, and finally used as interior components in passenger car doors. Despite extensive measures, the furnace and press still consume a lot of energy – contributing to a large ecological footprint for each sheet metal component. 

he remaining measures are less obvious. They require more information about the operational use of different forms of energy and resources, but offer potentially large savings. At a higher level, this means that data must first be collected, processed, and digitized, and monitoring systems must be set up. This is the only way to identify and exploit optimization opportunities in 
a structured manner. In operational terms, this means primary data instead of estimates, transparency at product level instead of average values, and continuous monitoring instead of one-off investigations. Companies need to link measured values to information about the systems, components, and products if they want to derive valuable insights. This increases the complexity of the evaluation and thus the manual effort immensely.

Digital twins are the solution. They link planning knowledge with real instance data and make the environmental impact of a specific product visible, not only in the laboratory, but on the line, per cycle, per part – in other words, for each individual instance. This allows companies to reduce production costs, mitigate the effects of potential price fluctuations, demonstrate their commitment to environmental sustainability, and meet regulatory obligations from energy management to the supply chain.

End of a shadowy existence?

Machines and systems are the heart of every factory and the source of data that makes products more sustainable. Energy consumption, resource use, waste streams, cycle time, and quality data are generated where value is created, i.e. with spindles, presses, furnaces, and robots. Companies must recognize the value of this data for the future of their manufacturing and learn to deal with the initially overwhelming amounts of data.

So far, only average values have been collected from the hot forming line for steel sheets. These include the daily energy consumption of the press, the set temperature of the furnace, and the average scrap rate. These values are insufficient for modern energy management methods, digital product passports (DPPs), and product carbon footprints (PCFs).

Moving forward, they will require more precise data, including details on individual machines, products, or material batches. This data must be clearly assigned to a serial product along its product life cycle, spanning from the planning phase (»as planned«) to the recycling phase (»as recycled«).

Digital twins connect these data streams. The »digital master« specifies the blueprint for each product instance using CAD / PLM data, as well as process and LCA models. The »digital shadow« is the specific dynamic or live data of each individual real product. The core of the twin never stops working. It updates the observed parameters for monitoring, visualizes hotspots, triggers decisions, and automatically optimizes process parameters with regard to energy and emissions. The result is a reliable, individual PCF for each product instance and the data basis for the DPP.

The medium-sized supplier has confirmed that a pilot project using digital twins of the furnace, handling robot, and sheet metal part can quantify PCF and other LCA categories in real time for each part. Rather than using average values, each digital shadow can be tracked individually. Analyzing instance data reveals levers for initiating sustainability effects, such as material selection, energy and media consumption, and scrap. In other words, these are precisely the aspects where machines and systems make a difference.

»EnerDiZ« sheds light on the darkness

Researchers at Fraunhofer IPK are developing a practical reference framework, a methodology, and a modular demonstrator for the targeted use of digital product and machine twins for energy efficiency with their EnerDiZ approach. The basic idea behind EnerDiZ is clear: energy and resource inputs are assigned to a specific product in addition to the annexes that produce it. This makes energy efficiency an optimizable product characteristic. The basis for this is a reference architecture that integrates production and annex data, for example on energy, cycles, or states, with product and process models as well as LCA / PCF calculations. Open standards like the Asset Administration Shell or AML make the solution scalable. A modular demonstrator maps typical manufacturing processes from machining to energy-intensive thermal or forming steps. This demonstrator will test data acquisition, mapping per part, event-based LCA calculations, and visualization in a practice-oriented scenario in the future. At the same time, the researchers measure potential savings, validate part and line solutions, and derive concrete recommendations for action, especially for SMEs.

In the future day-to-day business of our fictional supplier, EnerDiZ will initially collect the necessary data. Sensors in the press and furnace as well as the energy management and production control systems will provide events and consumption data. Each sheet metal part will be given a unique ID. Nothing escapes EnerDiZ, as every instance of a part is tracked from delivery of the sheets, to delivery of the finished component, and on to each individual car. Automated, event-based consumption mapping translates process signals into energy and material flows with instance-specific accuracy. Based on this data, an automated life cycle assessment updates the PCF and other impact categories of the finished vehicle component. Dashboards visualize hotspots, from materials to individual process steps to transporting the sheets.

The transparency provided by EnerDiZ helps to derive targeted measures. These measures include simple parameter adjustments, such as process temperatures, pressure, or standby strategies. They also include planning decisions on batch sizes and sequences and design feedback. AI methods such as reinforcement learning are integrated within secure guardrails to automatically find low-energy settings. Finally, the information is 
collected in such a way that it is available for DPP and PCF in a manner that complies with regulations. This primary data-based PCF / DPP data also facilitates audits and reports (ESPR, CSRD) and ensures compliance.

Into the spotlight

EnerDiZ uses instance-based data instead of averages, processing it transparently, and thus making the energy and environmental impact per product visible and controllable. The solution is transferable and scalable. Its modular structure and standard-based models enable transfer to other machines, lines, and factories. EnerDiZ builds a bridge to data ecosystems. It considers connectivity to infrastructures such as Catena-X from the outset and seamlessly supports future regulatory data flows. 

Feedback-to-design approaches allow factory knowledge to flow back into development and purchasing, making material and process decisions effective in the long term. SME are given an ideal entry point: The process model, guidelines, and laboratory demonstrator make it easy to get started and speed up implementation. EnerDiZ is paving the way from the heart of the factory – the machine – to the digital product twin and back.