Based on interviews in industry, this article describes a common understanding of the most important use cases. The results capture the current direction in the plastics industry.

Part 1 of the Series: Use Cases Based on Strategic Priorities in Plastics Processing

Strategie und Organisation

Wirtschaft

Spritzgießen

Kunststoffverarbeitung: Verfahren

Industrie 4.0 für die Kunststoffverarbeitung

Industrie 4.0

Compoundieren

Kunststoff-Aufbereitung

Implementation of Industry 4.0 in Injection Molding and Compounding

The new head of the Asia-Pacific region (excluding Greater China) will be responsible for sales, service and project management.

Spritzgießmaschinen, Spritzpressen

Kunststoffverarbeitungsmaschinen

Spritzgießmaschinen, allgemein

Kunststoffverarbeitung: Maschinen und Komponenten

Spritzgießmaschinen

Automatisierungstechnik

New Vice President of Asia-Pacific Region

A consistent flow of data and information is essential for digital solutions and meaningful analytics. The introduction of potential solutions is therefore often associated with high costs for data preparation.

Standardization Guarantees High Data Quality for Future Applications

Digitale Transformation

Innovationen und neue Technologie

Vernetzung / Cloudlösung für die Kunststoffverarbeitung

IoT-Plattformen für die Kunststoffverarbeitung

To Get the Information Flow in a Row

A year to forget is nearing its completion. To brighten up this dreary fall season without a major event like Fakuma, injection molding machine builders have concocted various strategies.

Injection Molding Machine Builders Are Discovering New Ways to Reach Their Customers

Biokunststoffe

Kunststoff Werkstoffe

Nachhaltigkeit

Umweltmanagement

Schneidmühlen

Silikonkautschuk

Kautschuk / Elastomere

What Can You Do at Fall Fairtime,  When There Aren’t Any?

Once again, the K 2019 in Düsseldorf, Germany, lived up to its claim to be the leading trade show for plastics technology. At least in the field of injection molding, there were exciting global innovations to see – if you looked closely.

Injection Molding Trends at K 2019

Schaumspritzgießen

Smart Factory

Energieeffizienz

Mikrospritzgießen

Mehrkomponenten-Spritzgießen

Spritzgießwerkzeuge

Werkzeuge für die Kunststoffverarbeitung

Fließpressen

Handling, Handhabungstechnik für die spanende Fertigung

Handling, Handhabungstechnik

Spritzgieß-Sonderverfahren

Spritzgießsimulation

Software

Simulation

Elastomer-Spritzgießen

Linearroboter für die Kunststoffverarbeitung

Industrieroboter

Montagetechnik

Montage-Spritzgießen

Knickarmroboter

Co-Injektion (Barriere-/Sandwich-Spritzgießen)

Hybrid-Spritzgießen (Metall / Kunststoff)

Ressourcenmanagement

Produktion

Technical Excellence thanks  to Digital Assistance

The resource-economical determination of an optimized operating point in injection molding is still a challenge. Machine learning offers great potential for re-using knowledge from a flow simulation that has already been generated.

Machine Learning: Combining Simulation and Experimental Test Data

DOE-Software

Qualitätsmethoden

Digitaler Zwilling

Quicker Evaluation of the Optimum Operating Point

In plastics processing, efficient production methods are required, especially for articles with high volumes and exacting quality requirements, and these must be constantly optimized. The container specialist Walther Faltsysteme has been successfully working together with the Austrian moldmaker Haidlmair for many years.

Walther Faltsysteme Combines Three Mold Technologies from the Haidlmair Group

Industrie 4.0 für die Qualitätssicherung

Condition Monitoring / Prozessüberwachung

Instandhaltung

Paletten und Behälter

Logistik

Verpackungstechnik

Dienstleistungen

Werkzeug-, Formen- und Modellbau

Heißkanalsysteme

Komponenten für die Kunststoffverarbeitung

With Broad Sword and Data Collector

The XPlanar motor system combines the advantages of conventional transport systems with magnetic levitation technology. This opens up completely new possibilities for handling and positioning products in the machining process. The floating effect obviates the need for mechanical guiding components and helps Plasmatreat to simplify the system configuration for plasma treatment.

XPlanar Transports Workpiece during Plasma Surface Treatment Process

Automatisierungskomponenten

Förderanlagen und -geräte

Oberflächentechnik

Fördersysteme

Nozzle in Protective Position

With the increasing digitalization of injection molding, new process parameters are coming under the spotlight. Engel, together with the Polymer Competence Center Leoben and Montan University Leoben, Austria, is using machine-learning methods to explore the potential of mold breathing for quality assurance in rubber injection molding. This research project was supported by the partner company SKF Sealing Solutions Austria.

Automatic Process Monitoring with iQ Clamp Control and Machine Learning

Steuerungen (Hardware und Software)

Mess-, Steuer- und Regelgeräte

Qualitätssicherung

Qualitätssicherung Kunststoff

Machine Learning für die Qualitätssicherung

Rubber Injection Molding with  Quality Recognition

iQ process observer, the new process monitoring platform from Engel monitors hundreds of parameters from the injection molding cell in the background.

A Process Monitoring Platform Monitors Several Hundred Parameters Simultaneously

Prozessvisualisierung

An Eye on the Process –  a Handle on Quality

In an online survey, representatives of the plastics industry formulated their expectations toward the Plastics Innovation Center 4.0 (PIC 4.0).

Part 5 of the Series: Practical Demonstrators for the Plastics Innovation Center 4.0

Messdatenerfassung und -verarbeitung

Steuern, Regeln

Prozessoptimierung

Prozessmanagement

Machine Learning

Datenanalyse

Kunststoff-Rezyklat

Focus on the User Benefit

After a year’s forced break of Fakuma, Wittmann Battenfeld is announcing the new servohydraulic SmartPlus machine as the highlight of this year’s trade show. It is claimed to be characterized by high economic and energy efficiency and repeat accuracy.

Messe Fakuma

kunststoffe-international.com

Spritzprägen

Injection Molding Innovations In-House, Live and in Color

iQ motion control from Engel now allows reliable early start combined with fully automatically optimized motion planning. The assistance system has been part of the standard kit of all new Engel viper-type linear robots since October 2021.

iQ Motion Control Brings the Injection-Molding Machine and Robot Closer together

A Production Cell with Fluid Movements

Over past months, many in-house workflows and development processes have slowed down and also changed. Nevertheless, time has not stood still, also not in injection molding technology. The debate and development have been shaped by two themes: digitalization and the circular economy. Below are some examples of developments and projects that underpin this trend.

Innovative Injection Molding in Times of Pandemic

Künstliche Intelligenz für die Qualitätssicherung

Folientechnik (IMD, IML, FIM, u.a.)

Schnecken und Plastifiziersysteme

Machine Learning / Künstliche Intelligenz für die spanende Fertigung

Industrie 4.0 für die spanende Fertigung

Kunststoff-Recycling

Lasermarkieren

Laserbearbeitungen

Inline- / Online-Messsysteme

Kreislaufwirtschaft

There Is More beyond the Horizon

At Fakuma 2021, iQ Weight Control will be unveiled with its new interface and expanded functionality. The new developments have been driven by feedback from customers. With its consistently user-centered development approach, Engel is one of the pioneers.

The Assistance System iQ Weight Control Promises Greater Clarity and New Functions

Produktentwicklung(sprozess)

Qualitätsmanagement

Focusing on the User

On the Way to the Smart Injection Molding Factory

Part 2 of the Series: Integrated Data Acquisition for the Injection Molding Production

Everyone is talking about digitalization and Industry 4.0. In practice, however, implementation is often difficult. The data-driven operation of an injection molding production is often still a dream. Only if the exact actual condition of the machines and peripheral equipment is known, work on optimizing the processes can be done. The Eastern Switzerland University of Applied Sciences describes the challenges that have to be overcome and shows possible solutions based on a self-operated manufacturing cell.

There are two main challenges in implementing Industry 4.0 today [1].

On the one hand, there is often a lack of awareness in practice, both for possible applications and for the necessity of a concrete step-by-step implementation. We are talking about the actual target orientation or also use case definition for the smart factory. This concretization must be oriented to the following questions:

<ul><li>How can I increase my value creation by using data?</li><li>How can I use my data? What is the goal?</li><li>What data do I need to implement a specific use case?</li></ul>

The awareness for possible applications can be created by the use case framework presented in <a ext="tango:{f7174548-8ad9-4017-b06b-55b8e552a559}">the last article of this series (Kunststoffe international 7/21)</a>. This framework was systematized and developed on the basis of realized use cases in practice [2].

On the other hand, learning from data requires a complete database. Some questions arise here as well:

<ul><li>Which signals do I need? Which signals are available at all?</li><li>What quality of data do I need? Is the data available in this required quality?</li><li>How do I get the data out of my machine?</li><li>How do I synchronize data from different machines and devices?</li></ul>

The database is often a major hurdle, as the data is simply not available or not in sufficient quality. In the field of plastics processing, for example, there are no suitable, industry-wide standardized specifications and protocols that can be integrated with a variety of different data sources and database systems without any problems. This lack of standardization has resulted in each machine providing data differently today. Fortunately, many standards and specifications are currently in development (e. g. OPC UA), so interoperability will improve in the future.

Investigations showed that a step-by-step approach to implementing Industry 4.0 in plastics processing is target-oriented. We start with the definition of the use case. Based on this, an initial model for prediction can be developed with a focus on this use case. The validated model can then be integrated into the IT landscape, which is then rolled out in the factory or the entire production network. This is also reflected in the developed model for the implementation of Industry 4.0 in the factory (Fig. 1).

Many expert organizations are currently also dealing with this topic and specifically with the standardization of data. For example, there are also recommendations from the VDI group on which signals should be recorded in the field of injection molding [3].

Newer machines offer increasingly better data availability, i. e. they also enable more and more high-frequency recording of the individual signals and their storage as time series curves. This enables in-depth process analyses, which are essential depending on the use case. For the cyclic export of process parameters, the Euromap 63 and Euromap 77 (OPC UA) interfaces also play an important role here.

At the Institute of Materials Technology and Plastics Processing (IWK), it was possible to set up a data acquisition system to which five different injection molding machines from various manufacturers and various peripheral devices have now been integrated. The recorded data can be visualized and evaluated directly.

A number of challenges had to be overcome in order to set up this type of data acquisition for machine learning. It became clear that device-specific solutions often had to be implemented so that the data could be exported and recorded in the desired quality. This is a hurdle for smaller manufacturing companies in particular, as such interdisciplinary know-how is often not available.

Another difficulty is the synchronization of data from different machines and devices (e. g. injection molding machine, temperature control units, ambient conditions). Each device usually works with its own time stamp. This is a common problem, especially when working with time series of different devices. This can be avoided by setting up a live system in which all data is recorded synchronously. The time stamp is assigned by the data acquisition system. However, this requires that all machines and devices provide the data live. This means that curve data, which for example is made available via the OPC UA interface at the end of the cycle as time series histories, is not suitable for this purpose. A direct query of strongly time dependent signals (e. g. injection pressure or screw position) via OPC UA in high frequency is not suitable due to the limited publishing interval of the interface (usually max. 200 to 500 ms). However, it is possible to query pre-calculated parameters (such as maximum or average values) at the end of each cycle, and these data can also be combined with curve data. Nevertheless, aggregation already results in a loss of information. Depending on what is to be learned from the data, this can of course be critical. Older machines, which do not support such interfaces and do not store raw data at all due to limited storage capacities, present an additional difficulty.

In addition to synchronization, the allocation of data from different pre or post-processes is also an issue. On the one hand, data acquisition differs between continuous processes, e. g. material drying, in which continuous time series are recorded, and discontinuous processes such as injection molding and, for example, welding as a follow-up process, in which the data are recorded for each cycle and at a defined sampling rate. On the other hand, the data from the pre-process steps must then also be able to be clearly assigned to the later data of the part. For injection molding, this means which material batch was processed and how long was the material used, to manufacture the molded part, pre-dried? This is not a simple question, which is why it is currently still being worked on. For follow-up processes, this is much easier, since the part and the associated data already exist and only need to be expanded.

Fully Automated and Self-Learning Manufacturing Cell

One implementation of this connected factory is a manufacturing cell for the production of a floorball (Title figure and Fig. 2). The ball halves are manufactured on an injection molding machine, laser marked, completely measured in three dimensions and then stored in an interim storage system sorted by color. A collaborative robot is used for automated handling.

For customer orders, this robot takes the ball halves in the desired color from the storage system and transfers them to the welding machine, where the halves are welded into a ball fully automatically. This is a fully connected and automated production cell, the process data of all production steps as well as 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, the entire production process can be optimized by using artificial intelligence and advancing towards zero-defect production. This manufacturing cell serves as a training object for students of the OST –Eastern Switzerland University of Applied Sciences and shows companies the possibilities of digitalization.

Use of Data Acquisition for  Research Projects

The data acquisition system also serves as the basis for the use of artificial intelligence in a wide range of projects in the field of injection molding. A central topic is the early detection of process anomalies on the basis of process data. When an anomaly occurs, the machine operator should also be given a recommendation for a suitable countermeasure. This topic is being investigated in a three-year publicly funded research project with five industrial partners.

The process anomalies can be reliably detected and first appropriate countermeasures can be suggested. With the use of a support vector machine and an autoencoder an anomaly score (Fig. 3, top left) can be calculated based on various features. This indicates how stable the current process is running. When this score increases, a recommendation for action is suggested to the operator (Fig. 3, bottom right), which can be used to correct the process.

Other use cases include the prediction of the quality characteristics of the manufactured parts on the basis of process data and predictive maintenance or detection of wear on tool or machine components.

[1] Hopmann, C. & Schmitz, M. (2021). Plastics industry 4.0: Potentials and applications in plastics technology. Hanser Publishers; Hanser Publications.

[2] Budde, Lukas, Hänggi, Roman and Friedli, Thomas. „Digitalisierung à la carte: Individuelle Auswahl von Use Cases nach strategischen Prioritäten“ Zeitschrift für wirtschaftlichen Fabrikbetrieb, vol. 113, no. 12, 2018, pp. 850–854. <a href="https://doi.org/10.3139/104.112016" target="_blank">https://doi.org/10.3139/104.112016</a>

[3] VDI-Gesellschaft Materials Engineering (2019). VDI-Statusreport Industrie 4.0 in Spritzgiessunternehmen.

OST Ostschweizer Fachhochschule Inst. f. Werkstofftechnik u. Kunststoffverarbeitung (IWK)

https://www.iwk.hsr.ch

Ausgabenartikel

Fully automated manufacturing cell for the production of a floorball at IWK’s technical laboratory in Rapperswil, Switzerland

16:9

Fig. 1. Model for the implementation of machine learning in the factory

Fig. 2. Production flow of fully automated manufacturing cell

Fig. 3. GUI (Graphical User Interface) for anomaly detection and countermeasure proposal 

Ostschweizer FH - Hauptartikel

Curdin Wick, M.Sc., is head of the injection molding department at the Institute of Materials Technology and Plastics Processing (IWK) at Eastern Switzerland University of Applied Sciences (OST).

Prof. Dr.-Ing. Frank Ehrig heads the Institute for Materials Technology and Plastics Processing (IWK) at OST.

Prof. Daniel Schwendemann is head of the compounding and extrusion department at the Institute of Materials Engineering and Plastics Processing (IWK) at OST.

Prof. Dr. Roman Hänggi is a partner at the Institute for Product Design, Development and Construction (IPEK) and Professor for Production Management at OST.

Adrian Rüedy, M.Sc., is project manager at the Institute for Product Design, Development and Construction (IPEK) at OST.

The last article in this series takes a look at the challenges involved in implementing specific use cases and learning from data with the help of artificial intelligence in the field of compounding production technology.