Prof. Dr. Wolfgang Schlüter
Head of Competence Centres Flow Simulation (KSTS) and Industrial Energy Efficiency (KIEff), University of Ansbach
Research & Innovation Equipped for the Future with a Digital Production Image
The industry supplier pressmetall Gunzenhausen GmbH in cooperation with the University of Ansbach introduced a new IT solution for a digital production image. Its customized dashboards significantly improve production planning and control. The solution lays the foundation for further digitization measures.
Research at the Competence Center Industrial Energy Efficiency (KIEff) at Ansbach University is striving to increase the efficiency of the production processes in non-ferrous melting and die casting plants. In various projects of the Green Factory Bavaria research association and the Nuremberg Energy Campus (EnCN), intelligent solutions for melting and die casting operations of tomorrow are being developed together with industrial partners. The focus is on simulation, data analysis, automation and smart services.
Digitalization in the SME Sector
Small and medium-sized enterprises are generally very hesitant in implementing digitalization. In a study conducted by the Fraunhofer IPA 1, half of the several hundred companies surveyed have not yet taken any concrete steps to identify and exploit the potential offered by intelligent network solutions. At the same time, less than 5 % of companies see too little benefit in the introduction of new production IT. One reason for the discrepancy between the great interest in digitalization and the small number of projects implemented is that it is difficult to express the benefits of an IT solution in figures and therefore a given return on investment (ROI) can rarely be guaranteed. However, pilot projects are an excellent way of tackling digitalization, as they can convey its benefits without significant financial expenses.
However, small and medium-sized enterprises often do not have enough IT knowledge to implement pilot projects. The best way to close the competence gap with large companies is to establish partner networks. In this way, digitalization projects can also be successfully implemented in medium-sized businesses, as has happened between pressmetall Gunzenhausen GmbH and the University of Ansbach.
Digitalization in Foundries
When the KIEff of the University of Ansbach looks at various medium-sized foundries, the picture is often the same: The transparency of production is limited to local displays on the machines, there are no tools for operational planning and control of production and there is still a long way to go before "paperless production" can be achieved. This is confirmed by the study "Digitalization in medium-sized businesses" conducted by the Fraunhofer Institute for Production Technology and Automation (IPA) 2, in which machine data for half of the SMEs from the metal and electrical industries examined are only transmitted locally to the workers via the display systems of the machines. Although ten of the twelve companies surveyed rate the importance of digitization as being most important, only three companies have anchored digitalization in their corporate strategy. In addition, on average less than one percent of sales is invested in digitalization. Since foundries are often suppliers that produce mass products without being close to the end customer, the focus of digitalization must be on improving quality and efficiency of production. Data-based assistance systems offer high saving potentials.
In general, the following similarities between the studies and the companies visited were found with regard to the digital maturity level in medium-sized foundries:
- Digitalization is not anchored in the corporate strategy,
- Minimal investment in digitalization projects,
- No use of online machine data for production planning and control,
- A large amount of machine data is collected, but rarely used,
- Machine data is not collected centrally,
- Machine data is transmitted to the worker only locally at the machines or via central display panels,
- Isolated solutions through Excel lists, ERP systems and various databases,
- Experience-based and non-data-based approach.
Due to the low degree of digitalization described above, it is sufficient to implement a continuous provision of production-relevant data and information in a first step. Further steps in the context of industry 4.0, such as individualized products or cyber-physical systems, are to be taken if necessary.
In 2016 pressmetall Gunzenhausen GmbH and the research institute KIEff of the University of Ansbach joined forces to create a digital representation of the production by using individual software in the smelting and die casting plant in Gunzenhausen.
In addition to optimizing the production process and improving delivery reliability, the aim was to understand the added value of digitalization, dispel skepticism and identify problems with the introduction of new IT solutions in medium-sized foundries.
The pilot project was essentially divided into two project phases. First, a centralized and structured acquisition of all machine data as well as a decentralized online visualization of the machine data were implemented. In the second project phase, specific use cases were identified together with the employees to improve the production process through online access to machine data. This has resulted in various views tailored to user groups, which primarily include the production-related indicators of availability, performance, quality and overall plant effectiveness (OEE).
Structured and Central Acquisition of Machine Data
Before the start of the project, the data infrastructure in the plant resembled a typical foundry, which Jung 3 describes as follows: Although a considerable amount of technical, logistical and commercial data is acquired, this is often done by means of isolated solutions. Depending on the company, there are databases offering production data in real time or only manually managed Excel lists. Order data and product types are in turn stored in the ERP system. This environment does not allow an effective data-based process optimization. Therefore, the experience-based procedure is still used. However, if the machines are digitally integrated, Jung sees great potential for the so-called Smart Factory in foundries by creating a digital twin.
At pressmetall's plant in Gunzenhausen, the machine data captured by sensors are currently available as raw data in various file formats on machine computers. The order data is captured in an ERP system. Raw data typically have a low quality with regard to consistency, redundancy, relevance, comprehensibility and availability. Due to the poor data quality, it was therefore necessary at the beginning of the project to store the raw data on a database server in aggregated form in order to obtain central and structured data collection with improved quality. The PMGConnect tool used for this is executed on the project computer installed for the pilot project and accesses the various machine computers and the ERP system (Fig. 1).
In addition to the centralized collection of the machine data, the PMGOnline tool was introduced during the first project phase to visualize the machine data online. During the development the tool, the highest possible priority was given to operational independence and flexible configuration of the visualization. It was extremely important for PMGConnect and PMGOnline to create a virtual development environment with SimPMG that would enable software development independent of location and detached from the real production process.
The acquisition and visualization of machine data enabled data analysis and a decentralized overview of production and critical operating situations (Fig. 2), but the mere transparency of production did not yet allow sufficient use cases to be identified by the users. After a feedback round, a more specific adaptation of PMGOnline to the needs of pressmetall was aimed at, taking into account a significantly larger number of use cases.
Focus on the Production Area
The fundamental goal of pressmetall in the application of PMGOnline is to improve delivery reliability. This is to be achieved by enabling the user to determine whether the production target is jeopardized and the cause of a possible hazard. Possible causes for a threat to the production target may lie in low availability, performance or quality factors of the die casting machines. Altogether, these factors represent the overall system effectiveness (OEE). The individual factors can be assigned to the maintenance, foundry and quality departments. Further user groups in the production area are process technologists, production managers (smelters and foundries) and other production personnel such as forklift drivers or machine operators. An online representation of the production process allows a quick assignment of the cause of a hazard, which accelerates the process of finding a solution. The following user-specific views have been developed:
- General view of the plant: It provides a quick overview of the operating status of the machines (Fig. 2).
- Total view according to machine type: Tabular overview of the foundry or smeltery.
- Detailed view of the machine: Current and historical parameter values of an individual machine visualized in graphs or tables (Fig. 3).
- Dashboard die casting machine: It shows the current values and the history of production-related parameters and key figures of a die casting machine.
- Process stability die casting machines: A timeline assigns colors to visualize faults that occur in each die casting machine (Fig. 4).
- Overview KPIs: Production-related key figures such as availability, performance, quality and OEE are visualized in tables for each machine and highlighted in colors.
- Standstill overview: An overview of all idle machines.
Implementation of the Software Package
Within the framework of a pilot project, a swift and cost-effective provision of functionalities is required. The ODBC database interface was used for data acquisition, since most of the data was already available in databases. Modern technologies such as OPC UA or MQTT therefore did not offer any decisive advantages in this special case. The PMGOnline tool was implemented as a desktop application in the Matlab programming environment according to the client database server architecture. If the tool is used outside the pilot project, migration as a web application is to be implemented in the future. If further IT solutions are to be used in production in the future, thus resulting in a larger application landscape, a service-oriented architecture is to be preferred in order to promote reusability, openness and changeability.
PMGOnline has been implemented in a modular manner throughout the system, so that the views displayed can be flexibly configured by the client in Excel. This also allows the parameter and machine scope of the views to be individually configured. The modular software architecture also makes it possible to add new views in the future without having to adapt the software core.
Improved Production Planning and Control
With PMGOnline, the planning, control and efficiency of the production process at pressmetall's production site in Gunzenhausen could be considerably improved. The tool is currently installed on the workstations of foundry managers, smelting managers, process technologists, forklift drivers as well as maintenance and other production personnel. The tool is also used for production planning during shift meetings, which take place in a large group in a meeting room before each shift. This represents a significant improvement compared to the previous situation, in which magnetic boards and manually calculated key figures were still used.
In less than one man-year, a tool could be created without major financial expenditure, which is currently being used profitably in everyday production at pressmetall. The benefits generated by the pilot project, coupled with its low cost, show that digitalization can also be carried out in small steps and does not necessarily require the introduction of a comprehensive IoT platform. As an additional effect, the pilot project has given further impetus to employees who were open to digitalization. Various thought-provoking impulses and ideas have emerged as to how IT solutions could improve the internal production process. The pilot project was also an incentive to push ahead with digitalization.
Problems of Acceptance
During the pilot project, PMGOnline created a digital representation of the production process. Due to the lack of clarity in the data acquisition, it did not always correspond exactly to reality. This is not very surprising under real production conditions. However, employees who have a critical view of digitalization or who fear that their jobs are threatened by it, often use flaws in IT solutions as an opportunity to deny its benefits. Additionally, older employees are trying to maintain their familiar work patterns. The younger staff, on the other hand, are used to apps that function at any time, regardless of the platform. They often lack an understanding of the efforts involved in development, test phase and implementation of an individual software. These age-specific conflict potentials were frequently observed during the implementation of the pilot project. Digitalization is therefore not only a technological challenge, but also often requires a great deal of persuasion in order to integrate the new solutions into the day-to-day work of the employees.
IoT Platforms on the Rise
Current market trends show that IoT platforms are clearly gaining ground. They enable the networking and interaction of sensors, machines, devices and applications in the Internet of Things. This represents the technical and business foundation for data-based business models. According to a survey of 533 industrial companies conducted by the Digitalverband Bitkom e.V. in 2018, eight out of ten companies are already discussing the use of IoT platforms, while four out of ten are already using them 4.
According to another study by the IT market researchers of the International Data Corporation (IDC), 51% of 444 organizations in 8 focus industries plan to introduce an IoT platform in 2018. In manufacturing, the proportion of companies that have allocated budgets and plan IoT projects this year is 67% 5.
The studies show that it will only be a matter of time before IoT platforms also find their way into medium-sized foundries. Compared to large companies, SMEs tend to rely on external IoT platforms because they have less IT knowledge and a smaller number of users. IoT platforms would not only cover the implemented functional scope of tools such as PMGConnect and PMGOnline but would also significantly expand and simplify them. By supporting various interfaces, the machine park could be connected without having to completely program the structured data acquisition itself. Applications can be purchased through Software-as-a-Service (SaaS) solutions or developed in Plattform-as-a-Service (PaaS) solutions.
The software package described here enables efficient production planning and control up to the introduction of an IoT platform at pressmetall. The usage of PMGOnline, created an incentive to move to an IoT platform supported infrastructure. The identified use cases from PMGOnline can be used for the introduction of an IoT platform. If the software package is continuously updated as part of the research project, it is also possible that the present work will be the starting point for a new industry-specific IoT solution.
This article was first published by GIESSEREI.
- 1) Fraunhofer Institute for Manufacturing Engineering and Automation, Industrie 4.0: Entwicklungsfelder für den Mittelstand, T. Bauernhansl (Ed.), Stuttgart: kom|werb Agentur, 2016.
- 2) Fraunhofer Institute for Manufacturing Engineering and Automation: Digitalisierung im Mittelstand - Entscheidungsgrundlagen und Handlungsempfehlungen, T. Bauernhansl (Ed.), Stuttgart: kom|werb Agentur, 2017.
- 3) Giesserei 103 (2016), [No. 12], pp. 50-51.
- 4) Bitkom e.V., IoT platforms are a topic in eight out of ten companies, 22.05.2018. Online] Available at: https://www.i40-magazin.de/newsarchiv/iot-plattformen-sind-in-acht-von-zehn-unternehmen-thema/(date of access 20.09.2018)
- 5) International Data Corporation (IDC), "After major start-up difficulties, 2018 may be the year of the IoT breakthrough for companies in Germany," March 22, 2018. Online] Available at: https://idc.de/de/ueber-idc/press-center/65730-idc-studie-nach-gro-en-anlaufschwierigkeiten-kann-2018-das-jahr-des-iot-durchbruchs-fur-unternehmen-in-deutschland-werden/ (date of access 06.11.2018).