Digitalization as a lever for recycling and plastics production

28.05.2026

How can the plastics industry meet the growing demands for sustainability, efficiency and competitiveness? The third webinar on Friday in our series "Transformation. In dialog / Materials of the future" with two perspectives that are closely linked: Dr.-Ing. Katharina Krause, an expert in sustainability and the circular economy at Neue Materialien Bayreuth GmbH, shed light on the role of digitalization in plastics recycling from a scientific perspective. Fabian Landmann from seitcom GmbH then showed how artificial intelligence in plastics production can make a concrete contribution to creating transparency, improving planning processes and leveraging efficiency potential.

Rethinking recycling: Why digitalization is becoming a key competence

Dr.-Ing. Katharina Krause showed that recycling today is much more than a question of disposal or technical processing. Rather, it is developing into a strategic field of innovation that is significantly influenced by regulation, climate targets and resource pressure. Increasing demands on recyclability and the use of recyclates as well as the need to replace fossil raw materials are fundamentally changing industrial practice.

It is not enough to simply recycle materials. It is crucial to create high-quality cycles in which materials are recycled in such a way that they can be used again in demanding applications. It is precisely at this point that the complexity of plastics recycling becomes clear. Different additives, coatings, manufacturing processes, ageing conditions or contamination ensure that material flows are rarely homogeneous in practice. Even materials made from the same polymer can differ greatly in terms of their processability and subsequent application potential.

It is precisely this heterogeneity that makes the selection of suitable recycling processes, stable process management and quality assurance difficult. Mechanical and chemical recycling are not opposites, but different technological paths that must be used sensibly depending on the material flow. The decisive question is not which process is better across the board, but which is best suited to the respective material flow in technical, ecological and economic terms.

In order to make these decisions on a sound basis, data is required. And this is exactly where digitalization comes in: It creates transparency along the entire process chain - from input analysis to laboratory and process data to the evaluation of quality, energy consumption and CO₂ footprint. It is only through this systematic use of data that a complex recycling process becomes a controllable system.

The look at data-driven methods such as machine learning was particularly exciting. Using examples from compounding and extrusion, Dr. Krause showed how processes can not only be optimized, but also increasingly predicted. This turns classic trial and error into a learning, predictive approach.

This made it clear that digitalization is not an end in itself in recycling. It is the prerequisite for developing economically viable, sustainable material cycles from complex and fluctuating material flows.

AI in production: A lot of data has long been available

Fabian Landmann turned his attention to production practice. His key message: the foundations for digitalization are already in place in many companies. It is often not the data that is lacking, but its structured use.

Typical data sources include MES data, PDA data, material consumption, scrap values, ERP information or quality data from quality assurance. This is an important realization for many companies, as it shows that getting started with digital applications does not necessarily require a major project.

It is therefore possible to get started pragmatically - for example with QR codes, tablets or simple data capture solutions. Even in paper-based environments, the first data collection points can be integrated, for example at raw material receipt, QA approvals or at the exit of finished products. Such "low hanging fruits" create initial transparency and form the basis for further digital processes.

However, classic planning methods quickly reach their limits, especially in complex batch processes, such as those commonly found in the plastics, chemical or pharmaceutical industries. If production planning is carried out using Excel, paper or implicit empirical knowledge, inefficiencies often remain invisible for a long time. Short-term downtimes, changing priorities and numerous dependencies make truly reliable detailed planning difficult.

This is exactly where AI comes into play.

Production planning with AI: more transparency, faster response, better decisions

Fabian Lattmann used specific practical examples to show how AI-supported production planning can work. Such systems dock onto existing ERP structures, take over order data and link it with resource, process and machine data. On this basis, production plans can be optimized in a short space of time and dynamically adapted as required.

This applies not only to regular planning, but also to ad hoc situations: If employees call in sick, machines break down or maintenance measures are due, an intelligent planning system can re-plan and run through scenarios within a short space of time.

This fundamentally changes the nature of planning: static planning becomes a dynamic, learning process. Companies can weigh up different targets against each other - such as adherence to deadlines, minimizing set-up times, energy efficiency or operating costs.

The transparency gained is also a significant added value. When target and actual data are compared with each other, deviations can be detected in real time. This not only improves the speed of response, but also the traceability of processes and the quality of decisions.

This makes it clear that AI does not replace the experience of employees, but complements it where the complexity becomes too high to keep an eye on all influencing factors at the same time.

Data security as a prerequisite for trust

Another point that resonated in both presentations was the topic of data security. In the plastics industry in particular, recipes, process parameters and production expertise are among the most sensitive corporate assets. Anyone who wants to use AI in this environment must therefore not only pay attention to functionality, but also to security and control.

Local hosting models or in-house data center structures can create important prerequisites for ensuring data protection, GDPR compliance and control over sensitive production data.

Digitalization is not an end in itself

Despite their different focuses, both presentations led to a common core statement: digitalization is not an end in itself in the plastics industry. It is a tool to make complexity manageable - in recycling as well as in production.

In recycling, it makes it possible to better understand heterogeneous material flows, select suitable processes and control processes in an ecologically and economically sensible way. In production, it creates transparency, improves planning and helps to respond more quickly and effectively to disruptions.

The realization that many companies do not have to start from scratch was particularly encouraging. The necessary data, processes and starting points are often already available. The key is to use these systematically and develop concrete applications with added value from them.

Our webinar impressively demonstrated that digitalization is becoming a game changer in the plastics industry on several levels. It helps to advance the circular economy both technically and economically and makes production processes more transparent, flexible and efficient.

The key takeaway from both presentations is that those who take a pragmatic approach to digitalization and combine it with real-life use cases can already make tangible progress today. In recycling, this means better decisions along complex material flows. In production, it means more transparency, greater responsiveness and more intelligent production planning.

The result is a plastics industry that is becoming more sustainable, more resilient and more competitive - not through digitalization for its own sake, but through its targeted use where it provides the greatest benefit.