Digitalisation and AI-based tools are becoming increasingly important for companies involved in plastics recycling. As Erema representatives emphasise, the growing complexity of processes and the need to maintain consistent recyclate quality while dealing with variable input material increase the demands on operators and maintenance teams. Another factor is the shortage of skilled labour, which means not every production shift has specialists available to assess the process continuously and respond quickly to deviations.
In the interview, Markus Huber-Lindinger, Managing Director at Erema, and Yvonne Kappacher-Winter, R&D Engineer at Erema, point to the role of digital solutions in increasing process transparency and supporting operational decision-making. Analysing process data is intended to make correlations visible that remain hidden in day-to-day operations. Erema refers to applications including automated quality and process monitoring and predictive maintenance based on real-time machine data, aimed at reducing unplanned downtime and improving production stability.
The company links the acceleration of digitalisation in recycling to simultaneous pressure for consistent recyclate quality and cost-efficient production. It also points to rising regulatory requirements and the expectations of global brands seeking higher recycled content. In Erema’s view, digitalisation can help compensate for fluctuations in feedstock, improve process stability, and reduce energy and resource consumption. Huber-Lindinger describes two parallel directions, digitalising the company’s internal production processes and advancing machine “intelligence”, including solutions designed to increase availability and support quality assurance.
Erema also highlights an approach covering the entire value chain. Together with Lindner Washtech, the company is developing integrated, data-driven process solutions intended to enable automatic interventions and condition-based maintenance. The stated goal is to improve transparency and efficiency across the entire route, “from bale to pellet”.
Key elements of the digital ecosystem
Among the solutions that Erema says currently create the most value for customers, three areas are highlighted. The first is predictive maintenance. The PredictOn tool is intended to enable real-time monitoring of key components such as main drive trains and the plastification unit. The system is designed to detect anomalies early and recommend specific service actions, which is expected to reduce unplanned downtime and extend component service life.
The second area concerns process and quality transparency. Erema points to the BluPort platform as a hub that brings together digital services, documentation and machine information, as well as data on current production and on-site machine performance.
The third area is automation in critical process steps, particularly where feedstock variability makes stable operation difficult. An example cited by the company is the DischargePro system for the Erema Laserfilter, which is intended to automatically adjust the speed of the scraper disc and the discharge screw to current demand. According to Erema, this supports uniform thickening of the melt during filtration.
PredictOn as an intelligent assistance system
Yvonne Kappacher-Winter describes PredictOn as an intelligent assistance system that accompanies the recycling process in real time. It is designed to continuously monitor and analyse relevant operating status data, identify patterns, and detect anomalies and deviations from normal operation. Early detection of wear on key components is intended to minimise the risk of unplanned downtime.
As Erema states, the aim is to maximise machine availability and ensure process stability. The system is intended to show operators when intervention is required, and maintenance actions are to be performed on a condition-based basis rather than routinely. The company believes this helps reduce unnecessary maintenance work while lowering the risk of costly failures. PredictOn is also intended to support teams in carrying out unscheduled maintenance activities in a targeted and well-coordinated manner.
Erema currently points to two modules. PredictOn:Drive is intended to monitor the main drive trains of the preconditioning unit and the extruder, as well as drive trains on the reactor, the crystallisation dryer and the vacuum pumps on PET systems. PredictOn:Plastification Unit, in turn, is intended to monitor the condition of the extruder screw and barrel in real time, without direct contact with the melt.
Process data, AI and deployment requirements
In Kappacher-Winter’s view, developing predictive maintenance tools for diverse applications is a complex task. Recycling systems process many types of materials, and machines differ in configuration, which makes data analysis more demanding and requires models that work reliably in different environments. Another challenge mentioned is coordinating internal collaboration across departments, from R&D and software development to service specialists.
Huber-Lindinger points to the role of AI-based algorithms in deeper data analysis within PredictOn:Drive, where they are intended to enhance pattern recognition and fault diagnostics. At the same time, he indicates the next development step, enabling AI applications to run directly on standard or retrofitted PLC-based control systems. The statement highlights the specifics of industrial applications, which must operate on smaller, specialised datasets, and the longer-term direction toward self-optimising machines.
Lessons from day-to-day use of digital tools
With regard to practical day-to-day operation, Kappacher-Winter emphasises the importance of simplicity. She cites an earlier deployment of smart glasses for commissioning support, which in practice were rarely used. Customers reportedly preferred smartphones to share photos and videos with service teams and to hold live calls directly at the machine. At the same time, she notes that more capable tools, such as training simulators, may increase the effectiveness of AR solutions in the future.
Interoperability is also described as a key implementation factor. Erema points to standards such as OPC UA, which are intended to ensure reliable communication between different devices and systems, which is essential for integrated recycling processes and industry collaboration.
Development outlook
In Huber-Lindinger’s view, customers can expect further automation in key process steps, expanded predictive maintenance functions and deeper integration along the recycling chain. In the coming years, digital solutions are expected to increasingly merge into holistic systems. Erema also expects continued expansion of cloud-based systems, which are intended to play a growing role in linking machine data, digital services and analysis tools across the recycling process.
