KIMBA - KI-basierte Prozesssteuerung und automatisiertes Qualitätsmanagement im Recycling von Bau- und Abbruchabfällen durch sensorbasiertes Inline-Monitoring von Korngrößenverteilungen

Die Baubranche gehört mit 587,4 Mio. t/a eingesetzten Gesteinskörnungen zu den ressourcenintensivsten Branchen Deutschlands. Durch Substitution von primären Gesteinskörnungen durch rezyklierte (RC) Gesteinskörnungen werden natürliche Ressourcen geschont und negative Umweltauswirkungen wie bspw. Treibhausgasemissionen um bis zu 85% reduziert. Bisher decken RC-Baustoffe mit 73,3 Mio. t/a lediglich 12,5 Ma.-% des Gesteinskörnungsbedarfs ab. Ihr Einsatz beschränkt sich mit 53,9 Mio. t/a (73,5 Ma.-%) bisher überwiegend auf Tiefbauanwendungen.

Hierzu arbeitet der Lehrstuhl für International Production Engineering and Management (IPEM) der Universität Siegen gemeinsam mit dem Institut für Anthropogene Stoffkreisläufe (ANTS) der RWTH Aachen, dem Deutschen Forschungszentrum für Künstliche Intelligenz (DFKI) sowie den Unternehmen MAV Krefeld GmbH, Kleemann GmbH und Point 8 GmbH daran, die Qualitätssicherung und Akzeptanz von recycelten Gesteinskörnungen (RC-Baustoffen) im Bauwesen zu verbessern. Mittels innovativer Sensorik soll die KGV-Analyse in Aufbereitungsanlagen automatisiert werden. Bildgebende Sensoren vermessen dabei das RC-Material bereits während der Aufbereitung, und mithilfe von Deep-Learning-Algorithmen werden Partikel segmentiert, Korngrößen vorhergesagt und digital aggregiert. Diese automatisierte Überwachung ermöglicht es, Qualitätsveränderungen frühzeitig zu erkennen und Aufbereitungsprozesse anzupassen.

Vonseiten des IPEM-Lehrstuhls ist unser Gruppenleiter Maximilian Lutz für die operative Projektleitung verantwortlich.

AGASTIK - Automated generation of assembly sequences and times from implicit knowledge

Nowadays, the creation of assembly sequence plans still requires an enormous amount of manual effort based on the expert knowledge of employees. Often, planning and target times as well as process sequences have to be estimated imprecisely based on empirical values and corrected manually during production. These manual activities result in major challenges and resource losses, particularly in pre-series production and the production start-up of complex assembly systems. Demographic change is also increasing the need to efficiently train new employees in the area of assembly. Employees in SMEs that produce small batches are particularly affected by these challenges and were therefore the focus of the project.

This is precisely where the AGASTIK research project came in, with the aim of reducing manual effort and improving planning quality at the start of production. To this end, assembly planning data was to be generated automatically. A human-centric software system was therefore developed within the project that uses artificial intelligence (AI) to analyze assembly processes recorded once with minimal effort and explicates the implicit knowledge of the employees that can be recognized in them. The explicit knowledge is then made available as assembly instructions in a user-friendly interface. The software system architecture comprises three main parts: 1. one-time recording of assembly processes, 2. automated analysis using AI and 3. presentation of the assembly instructions.

In addition to IPEM, the Machine Tool Laboratory WZL at RWTH Aachen University and the industrial companies Wilhelm Vogel GmbH Antriebstechnik, Medenus Gas-Druckregeltechnik GmbH and Weber Maschinenbau GmbH are also involved in the research project. The consortium is complemented by the two software developers Mobile Software AG and Lorenz Software GmbH. On the part of the IPEM Chair, our group leader Maximilien Schütz
has taken over the operational project management.

Funded by: Federal Ministry of Education and Research

EnEWA - Saving energy in paper production by tapping into the value chains of waste paper from lightweight packaging, residual waste and commercial waste

In order to reduce CO2 emissions in the production of new paper, make production more sustainable and optimize recycling, the EnEWA research project is pursuing the approach of tapping into new sources for environmentally friendly waste paper recycling and making them usable. To this end, the Chair of International Production Engineering and Management (IPEM), together with the Institute for Anthropogenic Material Cycles (ANTS) at RWTH Aachen University and the industrial partners Tomra Sorting GmbH, STADLER Anlagenbau GmbH and PROPAKMA GmbH as well as the paper producer LEIPA Group GmbH, is developing a solution for recycling waste paper from the lightweight packaging, residual waste and commercial waste value chains.

To this end, the entire process of waste paper collection, extraction and processing is being considered. Following an analysis of the types of waste paper and waste paper composites to be recovered and their quality as part of material flow analyses, the necessary sorting technology is to be developed and adapted. This will be followed by the development of a process for the hygienization and preparation of the paper fibres. On behalf of the IPEM Chair at the University of Siegen, our senior engineer René Sauerhas taken over the operational project management.
has taken over the operational project management.

Supported by: BMWK - Federal Ministry of Economics and Climate Protection

ventUS - Business Venturing at the University of Siegen

The EXIST program ventUS, initiated by the University of Siegen, has been successfully completed. The program, financed by the Federal Ministry of Economics and Climate Protection, aimed to identify innovative ideas at university chairs, laboratories and institutes earlier and more systematically than before. It offered numerous students and employees the opportunity to take the plunge into founding a company. An interdisciplinary team of scouts, coaches and advisors worked closely with the university's start-up office and Startpunkt57 to support participants in various phases of their start-up projects. As part of the ventUS project, founders benefited from extensive programme formats, tailored lectures and specialized events. These offerings enabled them to further develop their start-ups with professional support, receive valuable feedback and build important network structures.

The incubator program, in which the IPEM chair played a central role, deserves special mention. In jointly organized events and workshops, we supported those interested in founding a company from the initial idea to the concrete business model. Building on the incubator program, the transition to the accelerator program followed, which provided support for the actual start-up. In addition, the founders in these programs were effectively networked with investors, SMEs and other university members. An established coworking space complemented the broad range of support on offer.

The vision for the year 2030 of establishing the University of Siegen as a start-up hotspot in a model region in Germany, in which there is a sustainable symbiosis of innovative start-ups and successful SMEs, was significantly advanced by the ventUS project.

smart.CONSERVE - Smart Container Services for Food Industries

The food industry is subject to high requirements and strict standards in order to offer the end consumer consistently good quality and food safety. However, the seamless monitoring of limit values during the transportation and storage of foodstuffs proves to be challenging. For perishable foods in particular, it is extremely important to constantly track the defined parameters, such as the internal temperature or filling pressure of the stainless steel intermediate bulk container (IBC) used for transportation and storage, in order to detect deviations at an early stage and take countermeasures.

In order to meet these challenges, the research project "smart.CONSERVE - Smart Container Services for Food Industries" has seen the Machine Tool Laboratory WZL of RWTH Aachen University as consortium leader, the Chair of International Production Engineering and Management (IPEM) of the University of Siegen and the industrial partners Zentis GmbH & Co. KG and Packwise GmbH jointly researched the integration of information and communication technologies in stainless steel IBCs. As part of the project, an intelligent food container was developed that can record and log all relevant status data of the container using sensor technology.

You can find the project results in the following explanatory videos:
https://protech.mb.uni-siegen.de/ipem/research/images/wzl_smartcap_eng_sub (1080p).mp4
https://protech.mb.uni-siegen.de/ipem/research/images/wzl_smartcap_ger_sub (1080p).mp4

The project was funded by the Federal Ministry of Food and Agriculture; Project Management Agency Federal Agency for Agriculture and Food. On the part of the IPEM Chair at the University of Siegen, our group leader Philipp Nettesheim
took over the project management.

5GROW - 5G Real-Time Optimization of Welding Processes

The 5GROW research project successfully used the new 5G telecommunications standard for AI-based real-time control in industrial applications during the course of the project. An automated welding process served as an example. In this project, artificial intelligence (AI) and computer vision were used to determine the optimum welding parameters when the welding gaps to be welded fluctuated due to dimensional deviations of the parts to be joined. Computer vision was used to measure the width of the welding gap. The data on the welding gap width determined by computer vision formed the basis for deriving the optimum welding speed using artificial intelligence. Furthermore, 5G was primarily used for real-time communication between the welding robot and the AI server, which ensures real-time adjustment of the welding parameters.

The main innovation is the combination of the future technologies 5G and AI to enable wireless real-time control of automated, robot-guided processes. 5G has thus become an enabler to improve the efficiency of automated processes and the quality of products.

The project was funded by the Ministry of Economic Affairs, Innovation, Digitalization and Energy of the State of North Rhine-Westphalia. On behalf of the IPEM Chair at the University of Siegen, our research associate Gerald Kolter took over the project management.

ODiWiP - Optimized material cycle in the paper industry

The ODiWiP research project has successfully developed and implemented a functional demonstrator of an AI-based assistance system for operating a paper machine. This assistance system offers an efficient way of monitoring the production process, detecting and rectifying quality problems at an early stage and preserving the decision-making authority of the machine operators. The self-optimization function and the ability to expand expertise ensure that the system is continuously improved and meets changing requirements.

This project shows the potential of AI in industry and highlights the importance of intelligent assistance systems for increasing efficiency and quality in complex production processes. The successful implementation of the demonstrator opens up the possibility for further development and adaptation of the assistance system. It is expected that this technology will play an increasingly important role in the paper manufacturing industry and beyond in order to optimize production processes and ensure the quality of the end products.

Due to the high relevance and the recognized need for further developments, a follow-up project, "KIBAPap - AI-based operator assistance system in the paper recycling loop", was initiated. It has been funded since 01.07.2023 as part of the BMWK's Industrial Bioeconomy funding measure. The aim of this new project, "KIBAPap", is to further develop the prototype from TRL 6 to a scalable solution (TRL 8). This progress should ensure that the prototype undergoes the necessary maturity and practical tests to ensure successful and sustainable commercial usability.

The ODiWiP project was funded by the Federal Ministry of Education and Research (BMBF) as part of the "Digtal GreenTech - Environmental Technology Meets Digitalization" measure of the "Natural.Digital.Sustainable" action plan. The action plan is part of the BMBF's "Research for Sustainability (FONA)" strategy. Our group leader Alexander Becher has taken over the project management from the IPEM Chair at the University of Siegen.

ManuBrain - Artificial intelligence for industrial SMEs

The ManuBrain project developed a universal, scalable and open platform for artificial intelligence applications in industrial SMEs. ManuBrain was part of the lead market competition IKT.NRW, in which eleven innovation projects were supported with around 16 million euros.

Only a few SMEs have so far used AI processes to analyze sensor data in production facilities. The research project therefore developed an AI platform that helps to unlock the potential of data analysis for specific industrial applications.

The ManuBrain platform offers a comprehensive solution for data connection to machines, data transport, data storage and an execution environment for AI methods. The design and development of the platform is based on previously formulated and successfully validated use cases. The aim of this pioneering project was to offer small and medium-sized manufacturing companies (SMEs) a plug-and-play solution that enables the exploratory use of artificial intelligence (AI) to optimize their production processes.

After successful implementation, the platform was tested and validated in several industrial use cases. The developed solution has proven to be suitable for a number of applications, including the detection of anomalies in machine data and the optimization of production processes.

The project was funded by the European Regional Development Fund. On the part of the IPEM Chair at the University of Siegen, our senior engineer Fabian Steinberg took over the operational project management.

Further information can be found on the ManuBrain project homepage

Sensing & Sensibility - Organizing human and non-human cooperation - the case of cyber production management

In the course of the ongoing digital transformation, everyday technologies will change fundamentally: they will become proactive, autonomous and increasingly inscrutable for humans. Using the example of production management, this research project investigates how collaboration between humans and algorithmic agents can and should be designed. Possible designs are examined with regard to three potentially competing goals: performance, satisfaction and accountability. To this end, the three applicants will work together on an exploratory study. In general, examples of different types of human-algorithm collaboration will be created and their impact on the efficiency and effectiveness of the outcome, the job satisfaction and well-being of the people involved, and societal and regulatory implications will be investigated. While production management serves as an example, the project will address broader questions of how human-algorithm collaboration can be designed between the priorities of industry, the:the workers:in and society as a whole.

The project was led by Prof. Dr. Carolin Gerlitz, Prof. Dr. Marc Hassenzahl and Prof. Dr.-Ing. On the part of the IPEM Chair at the University of Siegen, our research assistant Lili Wu took over the operational project management.

Further information can be found on the project homepage Sensing & Sensibility

ReLIFE - Adaptive remanufacturing for the life cycle optimization of networked capital goods

The aim of the ReLIFE research project is to increase resource efficiency by extending the life cycle of capital goods. To this end, the adaptive remanufacturing approach was successfully developed in the research project. This describes an adaptive maintenance strategy that determines the optimum time and scope of maintenance measures based on sensor evaluations from a technical, economic and ecological perspective. A key component of adaptive remanufacturing is a decision model that supports users in deciding on the optimum time and scope of maintenance and remanufacturing measures.

The innovative character of adaptive remanufacturing lies in the adaptive nature of the process in terms of both time and content. Based on the sensor-monitored wear condition of components, preventive remanufacturing measures tailored to the respective situation are proposed. The resulting guaranteed performance of the capital goods forms the basis for innovative business models to ensure long-term productivity.

The decision model was successfully implemented as part of a browser-based software application and connected to the demonstrator using data technology. In addition to visual condition monitoring, the scope of functions includes the selection and scheduling of suitable measures. In parallel, three product-service system-oriented business models were developed for the concept of adaptive remanufacturing. The economic validation of these models revealed a financial risk due to long-term payments, which is why the focus of the project was placed on the classic product-oriented model. With the project results achieved, it will be possible in future to transfer the broadly developed methodology to other products, sectors and industries for comprehensive market development. The increased resource efficiency will in turn ensure the long-term competitiveness of companies.

The project was funded by the German Federal Ministry of Education and Research. On behalf of the IPEM Chair at the University of Siegen, our research associate Marius Wiggerhas taken over the project management.
took over the project management.

Further information can be found on the project homepage ReLife

MAproFli - Multivariable automation decisions for volume- and product-flexible flow assembly

Traditionally, the optimum degree of automation is determined on the basis of comparative monetary calculations. However, a purely monetary evaluation is not sufficient: influencing variables such as product design, reliability, availability, quality improvements and, in particular, flexibility have far-reaching effects on the success of automation decisions. The aim of the MAproFli research project is therefore to develop a decision-making methodology for the optimum degree of automation in volume and product-flexible flow assembly based on multivariable criteria. First of all, the decision-making ability is to be improved by a context-relevant selection of decision factors and the necessary degree of adaptability is to be mapped. Based on this, a selection and evaluation of automation alternatives can be made. An improvement in the decision quality for automation alternatives is to be achieved by transferring findings from ergonomics, decision theory and proven methods of profitability analysis. Finally, the decision-making effort is to be reduced by developing a software solution based on the decision-making methodology. The area of investigation focuses on the flow assembly of medium to large series.

In order to validate and weight the decision factors determined to date, the IPEM chair has conducted an online study in collaboration with the Machine Tool Laboratory (WZL) at RWTH Aachen University, which is primarily aimed at experts from plant and assembly planning.