Industry 4.0 and Digitalization in the Plastics Industry
Information
Industry 4.0 covers a wide range of possible technologies and applications. This advancing digitalization opens up new opportunities for production companies. This talk will highlight these opportunities as well as the challenges involved in implementing Industry 4.0 in injection molding and compounding/extrusion production technologies. One challenge in digitalization is to identify so-called use cases. Small and medium-sized companies in particular find this difficult.
Definition of use cases [1]
How can the relevant use cases for a company be derived? A focused and targeted approach based on specific use cases has proven to be promising. The challenge lies in maintaining an overview of possible technologies and deriving their benefits for individual use cases for your own company. The structured approach in the search for suitable use cases enables companies to identify technologies for their production based on their strategic priorities. Based on the use case framework for the smart factory, the evaluation of suitable use cases is carried out in six steps:
- Set strategic priorities
- Select use cases
- Adapting use cases to specific context
- Evaluate impact on strategic goals
- Allocation of potential and complexity
- Implementation plan
Based on this approach, possible use cases can be identified and evaluated with regard to their implementation.
On the way to the smart injection molding factory [2]
Once a use case has been defined, the right database must be made available for learning from data. This also raises a number of questions:
- Which signals do I need? Which signals are available at all?
- What quality of data do I need? Is the data available in the required quality?
- How do I get the data out of my machine?
- How do I synchronize data from different machines and devices?
Many expert organizations are currently working on Industry 4.0, especially the standardization of data. Among other things, there are recommendations as to which signals should be recorded in injection molding. Newer machines offer increasingly better data availability, with increasingly high-frequency recording of individual signals and their storage as time series sequences.
A manufacturing cell to produce a floorball was implemented at the IWK. The ball halves are produced on an injection molding machine, laser-marked, measured completely three-dimensionally and then sorted by color and stored in an intermediate storage area. Automated handling is carried out by a collaborative robot. When customers place orders, this robot removes the ball halves in the desired color from the temporary storage area and transfers them to the welding machine, where the halves are welded together fully automatically to form a ball. This is a fully networked and automated production cell: the process data for all production steps and the quality characteristics are stored in the cloud and can be clearly assigned to each ball. On the one hand, this ensures complete traceability, and, on the other hand, the entire production process can be optimized and advanced towards zero-defect production through the use of artificial intelligence. This production cell serves as a training object in the education and training of students and shows companies the possibilities of digitalization.
The established data acquisition also serves as a basis for the early detection of process anomalies based on process data. If an anomaly occurs, the machine operator is also given a suggestion for a suitable countermeasure. Other use cases include predicting the quality characteristics of the manufactured components on the basis of process data and predictive maintenance, or the detection of wear on tool or machine components.
Digitalization in Compounding [3]
Data acquisition along the entire value chain from compounding to injection molding helps to establish correlations between component properties and the data information from both production processes. In contrast to discontinuous injection molding, compounding is a continuous process. The data can therefore not be assigned directly to a product via a cycle but must be recorded continuously and then assigned to different material batches. The challenge here is to assign the recorded process data to the granules produced in such a way that it can be used for further processing. The constant monitoring of the process and thus the early detection of anomalies (process fluctuations) is very possible.
The aim is now to record and evaluate this information from the compounder and its peripheral devices as well as other sensors. A central data acquisition system will be set up at the IWK to record and collate the data and store it using a common time stamp. Significant process fluctuations are to be recognized and suitable characteristic values can be formed without too much loss of information. In one project, the aim was to improve process understanding when adding water-based liquid fillers and to monitor process stability. A central and immensely important topic for the future is the optimization of the energy efficiency of the compounding plant, as this can only be optimized by networking the entire plant. The knowledge gained in the field of compounding can also be used for extrusion. Correlation with the end product is easier here. This is because in the extrusion area, the recorded data can be assigned to the extrudate via the running meters of the profile, for example.
