The idea of PLM is to be able to design the handling of a product from the idea to recycling. But even efficient producers sometimes leave many product phases unmoderated. On the positive side, this means that there is still a lot of potential to be tapped in this area - with digital twins, for example.

The climate crisis shows the global community that the transformation from a throwaway economy to a recycling economy is inevitable. Product Lifecycle Management(PLM) systems are still often used today in such a way that the derivation of a bill of materials to define the manufacturing of a product is the central functionality of the software. This is a process that does not support recycling. PLM software should not be understood primarily as a monolithic IT tool. Rather, it is a number of applications linked together by a semantic network. The cross-disciplinary approach is ensured by corresponding processes and a set of rules.

For the recycling of materials, it is above all important to be able to dismantle products easily and to make the parts available for reuse as sorted as possible. If products are analyzed in terms of their components, they consist exclusively (if software is defined as a part) of parts and joints. These are not implemented as independent objects in PLM systems: Parts and joints would still be linked into the higher-level assemblies with BOM relations in this process. Furthermore, a 'join' relation would have to be introduced that links the join to the joined parts. This would result in benefits for all processes further down the lifecycle (Figure 1). For manufacturing, the joints could be linked to the relevant manufacturing stations and the manufacturing process could be defined by the sequence of joints (serial and parallel). Maintenance of products is done either by replacing parts/devices or by modifications. Again, disconnecting and reconnecting joints is critical; after all, disconnecting joints is an essential step for recycling. During product modeling, information relevant for recycling can already be provided in the 'joint' object.

Some PLM manufacturers already offer 'digital twins'. The essential idea behind this is to provide a digital representation from the PLM environment for each product. This accompanies the life cycle of the product. The classic physical product structure including the 3D geometry is a minimal definition of a digital twin. This could also be built using the parts and joints method presented above. If these objects are still assigned to systems, there are advantages for the entire life cycle, e.g. from manufacturing, condition analysis, maintenance to end-of-life.

Both the repair instruction and the repair itself can be documented in the digital twin. This is important for devices as well as cable harnesses, since most repairs occur in these domains. Whether it makes sense to also store a copy of the digital twin in the product also depends on whether the product can securely access the centrally stored digital twin in order to synchronize with it.

This is where the Internet of Things comes in. Sensors already provide status information in many products. Lufthansa, for example, has been monitoring the engines of its aircraft for many years. This enables it to determine at an early stage whether an engine could become unreliable. This information can be used to determine when maintenance would make sense. Furthermore, it is now common practice to monitor highly stressed mechanical components with regard to material fatigue, for example in wind turbines or elevators.

As already explained, data on the stress and reliability of the product are collected during its operation. This data must be made available to the PLM environment in processed form in order to check and, if necessary, revise both the quality of the simulations and the input parameters for the simulation. The more simulation and reality match, the more accurate statements about the reliability of the product become. This also enables manufacturers to offer new business models in which the business basis is no longer the sale of products, but services such as the availability of an elevator per year. Today, new functions are often made available through software. Tesla, for example, plans to equip its cars with new functionalities through software over the course of their lives. The cars are therefore fitted with sensors that are only used to a limited extent or not at all when they are delivered. They will only provide additional functionality later with new software. A digital twin is desirable for this purpose. One prerequisite for this is to ensure that software and hardware in a product are compatible along its life cycle. Building this functionality into a PLM environment is not trivial, but many providers are in the process of implementing it.

Acceptance of the PLM environment by users often depends first and foremost on usability. Given the complexity and extensive functionality of today's PLM environments, it is difficult to make them simple. It is therefore often advantageous to focus on the requirements of user groups, to optimize them and, if possible, to automate them. In addition to user integration, system integration also plays a role. What began with a PLM-ERP coupling is now a company-wide data hub. These cross-divisional and cross-system information and data flows in combination with the IoT form the basis for approaches such as closed-loop engineering. Thus, a key for the future is to recognize the industrial trends that are useful for oneself, to pick them up and implement them in one's own PLM.