This contribution presents the deployment of a knowledge-based Digital Twin tailored for brewing operations, derived from a general-purpose ontology based on the ISA-88 standard and extended with general rules. The Digital Twin integrates real-time process data with embedded expert rules and energy-aware process information, thereby enabling adaptive decision-making. The pilot implementation at the cyber-physical brewery of the RIF Institute for Research and Transfer e.V. demonstrates how semantic knowledge representation, OPC Unified Architecture-based integration, and rule-based reasoning improve process transparency and operational decision-making during fermentation. The achieved results illustrate the potential of knowledge-driven Digital Twins for interpretable, sustainable, and transferable control in process industries.

This paper proposes a method for automatically analysing routings using production data in discrete manufacturing environments. By integrating process mining and machine learning techniques, the method analyses process flows and their homogeneity to structure production into product families and visualise the associated models. Traditionally, Group Technology has supported the identification of product families and their associated processes by enabling the organisation of production around homogeneous objects and resources. This task is essential for current state analysis, serves as a key prerequisite for optimising production processes, and forms the foundation for Lean techniques such as Value Stream Mapping, Pull Production, and Line Balancing. However, as production systems grow more complex, traditional Group Technology faces limitations. Concurrently, Industry 4.0 has introduced advanced digitalisation, increasing the availability of production data. Leveraging this data to generate actionable insights remains a challenge for many manufacturers. This research addresses this challenge by proposing a data-driven method that overcomes the constraints of traditional Group Technology. The method is empirically validated in a discrete manufacturing setting. The study contributes to the field of production planning and control by demonstrating a novel application of process mining and machine learning for automated routing analysis in modern industrial environments.

High-frequency time series data in manufacturing creates computational challenges for machine learning applications. This study compares four feature extraction methods (PAA, PCA, catch22, tsfresh) for classifying surface-based defects in screw connections. Using 12,500 tightening operations across eight defect classes, we evaluate these methods based on classification performance, computational efficiency, and memory usage. Results show tsfresh variants achieve the highest accuracy (up to 11

Artificial intelligence (AI) is increasingly employed to enhance quality assurance in hybrid assembly systems, where humans and machines collaborate to ensure process reliability and product integrity. However, the effective adoption of such systems depends not only on their trustworthiness—as defined by regulatory and technical frameworks like the Fraunhofer AI Assessment Catalog and the European AI Act—but also on the trust of users and developers who interact with them in daily operations. This study examines how these two perspectives diverge and converge in the context of AI-assisted quality assurance. In the first phase, expert interviews (N = 11) were conducted to explore domain-specific interpretations of trust and trustworthiness and to identify critical system attributes influencing trust in AI. Building on these insights, a Kano analysis (N = 42) was carried out to classify AI attributes into must-have, performance, or attractive features from the perspective of practitioners. The findings reveal a notable gap between compliance-driven notions of trustworthiness and the experiential trust expectations of users and developers. These insights offer actionable implications for the design, governance, and human-centered deployment of AI in industrial quality assurance.

As Digital Twins gain importance in mechanical and plant engineering, the need for structured, interoperable, and machine-interpretable product information becomes increasingly apparent. This development highlights a long-standing challenge. Technical Documentation, although an integral component of products and essential for legal compliance and knowledge transfer, is still predominantly created in static and insufficiently accessible formats. Such formats limit its contribution to Digital Twin implementations and, more broadly, its potential value within digital manufacturing ecosystems. To better understand this discrepancy and identify pathways for strengthening the role of Technical Documentation, an interview study with experts from mechanical and plant engineering was conducted. Using Kuckartz’s structuring qualitative content analysis, the study reveals key challenges and requirements related to enhancing Technical Documentation for Digital Twin integration and enabling new use cases for artificial intelligence. Above all, the results show that heterogeneous cultural and legal frameworks in global target markets and the difficulty of obtaining accurate information for creating and updating Technical Documentation are central barriers that necessitate a holistic organizational knowledge strategy as well as interorganizational standards and processes. Based on these findings, measures for the further development of Technical Documentation in Digital Twin contexts are derived and presented. Taken together, the contribution of this article is to form a structured basis for conceptualizing a reference process for Technical Documentation, which in turn constitutes an essential step toward establishing a standardized documentation scheme.

This paper expounds upon the implementation of Six Sigma methodology in conjunction with Industrial Internet of Things integration for the qualification of the copper alloy CuZn30 in Cold Gas Spraying additive manufacturing.The utilisation of Six Sigma's structured approach facilitates the optimisation of process parameters, the reduction of defects, and the assurance of consistent quality in the production of CuZn30 components.The development of an advanced IIoT infrastructure enables real-time in-situ data acquisition, integrating sensors and monitoring systems to capture critical quality metrics. The study demonstrates how in-situ process data from the IIoT Infrastructure enhances statistical process control and enables data-driven decision-making during production.Through a combination of rigorous quality control and digital infrastructure, this work improves the repeatability, reliability, and overall efficiency of the CGS process for non-ferrous materials like CuZn30.

This paper presents an innovative teaching approach designed to prepare students for the demands of the modern, digitalized working world. Moving beyond traditional lecture-based courses, this method emphasizes hands-on, practice-oriented learning in interdisciplinary teams. The core of this approach involves students working with real-world demonstrators a brewing lab and a photobioreactor reactor (CO2 sink) at TU Dortmund University. These serve as platforms for students to apply theoretical knowledge in practical settings, particularly in digitizing systems using Industrial Internet of Things (IIoT) sensors. Key components of the teaching method include iterative learning, adaptive problemsolving and interdisciplinary collaboration. This paper outlines the conceptualization of this teaching method, including didactic approaches, application-related challenges, and learning objectives. It also presents a two-stage validation process involving experts and students, with the goal of incorporating this method into the curriculum as a formal course. This teaching approach aims to equip students with the skills needed to navigate rapidly evolving technological landscapes and changing workplace requirements, bridging the gap between theoretical knowledge and practical application in data science and IIoT.

This paper presents an AI-based condition monitoring service for predictive maintenance of a key component in a cold gas spray metal additive manufacturing system. Following a CRISP-DM methodology, the model leverages operational data collected from a technology demonstrator at TU Dortmund University and is further validated with data from the system manufacturer, Impact Innovations. The approach identifies critical indicators of component degradation to enable proactive maintenance scheduling, achieving promising performance on a holdout test set (Accuracy = 0.99, F1-score = 0.96) as part of a proof-of-concept demonstrator with Technological Readiness Level 5 (TRL 5). By integrating explainability methods and uncertainty quantification, the paper improves the transparency and reliability of predictive maintenance decisions, resulting in an overall more trustworthy operation of the condition monitoring service. While the primary contribution lies in advancing intelligent condition monitoring strategies, this approach is also envisioned as a catalyst for broader sustainable manufacturing frameworks. Future work will explore its integration into circular economy applications, where trustworthy condition monitoring may serve as a driver for circularity.

This paper presents a comprehensive collection of industrial screw driving datasets designed to advance research in manufacturing process monitoring and quality control. The collection comprises six distinct datasets with over 34,000 individual screw driving operations conducted under controlled experimental conditions, capturing the multifaceted nature of screw driving processes in plastic components. Each dataset systematically investigates specific aspects: natural thread degradation patterns through repeated use (s01), variations in surface friction conditions including contamination and surface treatments (s02), diverse assembly faults with up to 27 error types (s03-s04), and fabrication parameter variations in both upper and lower workpieces through modified injection molding settings (s05-s06). We detail the standardized experimental setup used across all datasets, including hardware specifications, process phases, and data acquisition methods. The hierarchical data model preserves the temporal and operational structure of screw driving processes, facilitating both exploratory analysis and the development of machine learning models. To maximize accessibility, we provide dual access pathways: raw data through Zenodo with a persistent DOI, and a purpose-built Python library (PyScrew) that offers consistent interfaces for data loading, preprocessing, and integration with common analysis workflows. These datasets serve diverse research applications including anomaly detection, predictive maintenance, quality control system development, feature extraction methodology evaluation, and classification of specific error conditions. By addressing the scarcity of standardized, comprehensive datasets in industrial manufacturing, this collection enables reproducible research and fair comparison of analytical approaches in an area of growing importance for industrial automation.

This paper explores how data space-based condition monitoring can enhance circular economy strategies and support sustainability in industrial production, using the RIF Cyber Physical Brewhouse as a case study. Data spaces, characterized by sovereign, interoperable, and trust-based data exchange, enable secure collaboration across stakeholders in the value chain. Drawing on a literature review and qualitative interview study within the Factory-X research project, the paper investigates the relevance of data spaces for sustainability from an industrial engineering perspective. The conceptual framework shows how sharing timely operational and environmental data through a federated data space for Festo's industrial valve terminal can inform lifecycle decisions. Two complementary services are used: a RGrading service assesses circularity potential (Reuse, Recycle, Remanufacture, Refurbish), while a Product Carbon Footprint (PCF)-Management Service allocates the component’s PCF based on its estimated remaining useful life. A method for assigning the valve’s initial PCF to the brewing process is proposed, illustrating how data spaces can operationalize sustainability goals and help industrial engineers integrate environmental metrics into operational planning decisions.

The ergonomic configuration of processes is becoming increasingly important, especially considering the changing demographics and increasing shortage of skilled workers. Exoskeletons are widely discussed as a means of protecting employees from overstraining at the level of personal protective measures. The field of industrial exoskeletons research is still relatively new and has many unanswered questions. For example, there have not yet been sufficient studies on the influence of exoskeletons on the movements of employees. This publication discusses the effects of exoskeletons in manual processes. For this purpose, exemplary physical activities are carried out in a pilot study by a subject collective, whereby the tasks are executed with and without an exoskeleton. During the execution, a motion capturing system is used to record the movement data. Different back-supporting exoskeletons are taken into account in the study. The evaluation is based on the joint angles of the participants while performing tasks with and without exoskeletons. It is shown that the use of exoskeletons has a significant effect on the movement patterns, with a distinction made between rigid and soft support structures.

Sales forecasting in make-to-order (MTO) production is particularly challenging for small- and medium-sized enterprises (SMEs) due to high product customization, volatile demand, and limited historical data. This study evaluates the practical feasibility and accuracy of statistical and machine learning (ML) forecasting methods in MTO settings across three manufacturing sectors: electrical equipment, steel, and office supplies. A cross-industry benchmark assesses models such as ARIMA, Holt–Winters, Random Forest, LSTM, and Facebook Prophet. The evaluation considers error metrics (MAE, RMSE, and sMAPE) as well as implementation aspects like computational demand and interpretability. Special attention is given to data sensitivity and technical limitations typical in SMEs. The findings show that ML models perform well under high volatility and when enriched with external indicators, but they require significant expertise and resources. In contrast, simpler statistical methods offer robust performance in more stable or seasonal demand contexts and are better suited in certain cases. The study emphasizes the importance of transparency, usability, and trust in forecasting tools and offers actionable recommendations for selecting a suitable forecasting configuration based on context. By aligning technical capabilities with operational needs, this research supports more effective decision-making in data-constrained MTO environments.

Recent advances in Machine Learning have significantly improved anomaly detection in industrial screw driving operations. However, most existing approaches focus on binary classification of normal versus anomalous operations or employ unsupervised methods to detect novel patterns. This paper introduces a comprehensive dataset of screw driving operations encompassing 25 distinct error types and presents a multi-tiered analysis framework for error-specific classification. Our results demonstrate varying detectability across different error types and establish the feasibility of multi-class error detection in industrial settings. The complete dataset and analysis framework are made publicly available to support future research in manufacturing quality control.

Hexaglide parallel manipulators are characterized by high accuracy and dynamic performance, which makes them suitable for industrial high-precision assembly tasks such as placement of electronic THT components on printed circuit boards. In this paper we describe an assembly system that comprises a Hexaglide manipulator with vertical ball screws, moving printed circuit boards relative to stationary THT components. We evaluate the effects of the manufacturing tolerances of machine parts, such as bar length tolerance, ball screw axis position uncertainty, and ball screw axis orientation uncertainty, on Hexaglide end-effector pose accuracy using a geometric simulation study based on stochastic tolerance sampling. In the investigated configuration and under standard industrial tolerances, bar length inaccuracy and axis position uncertainty lead to significant position and rotation deviations for the Hexaglide end-effector in the horizontal plane that need to be compensated for by control algorithms to enable THT assembly using the Hexaglide prototype. The geometric simulation method applied in this paper can be used by designers of Hexaglide machines to study and evaluate different machine configurations.

Delivery times represent a key factor influencing the competitive advantage, as manufacturing companies strive for timely and reliable deliveries. As companies face multiple challenges involved with meeting established delivery dates, research on the accurate estimation of delivery dates has been source of interest for decades. In recent years, the use of machine learning techniques in the field of production planning and control has unlocked new opportunities, in both academia and industry practice. In fact, with the increased availability of data across various levels of manufacturing companies, machine learning techniques offer the opportunity to gain valuable and accurate insights about production processes. However, machine learning-based approaches for the prediction of delivery dates have not received sufficient attention. Thus, this study aims to investigate the ability of machine learning to predict delivery dates early in the ordering process, and what type of information is required to obtain accurate predictions. Based on the data provided by two separate manufacturing companies, this paper presents a machine learning-based approach for predicting delivery times as soon as a request for an offer is received considering the desired customer delivery date as a feature.

The convergence of information technology with production technology, coupled with the escalating intricacy of production processes and products, is leading to new demands on the education of future engineers in mechanical engineering and related disciplines. New job profiles are arising and require interdisciplinary cooperation in the fields of information technology and the Internet of Things, machine learning and domain knowledge in order to enable data-based decisions for monitoring and improving products and processes in industrial production. The core of competence building has a strong focus on applying theoretical knowledge and information in order to make it tangible and thus enable people to learn. The contribution shows the conception and implementation of transnationally connected cyber-physical brewing labs, which were set up as learning factories for students and industry partners at TU Dortmund University and the University of Technology Sydney. The focus is on Industry 4.0 technologies, such as shared data space, condition monitoring of machines and assets in the Internet of Things, and the application of machine learning for product and process optimization. This article discusses the derivation of competence profiles, roles and the development of targeted theoretical and practical learning modules. It provides an overview of the use in various formats at both sites. The evolution of the brewing labs in existing research activities is also discussed. Finally, an outlook on future activities is given.

Learning from demonstration is one of the most promising methods to counteract the challenging long-term trends in repetitive industrial assembly. It offers not only a programming technique that is accessible to workers on the shop floor, reducing the need for robot experts and the associated costs but also a possible solution to the observable shift from mass-production to mass-customisation through flexible and generalising systems. Since the emergence of the learning from demonstration idea in the 1980s, its methodologies, capabilities, and achievements have constantly evolved. However, despite reports of continued progress in academic publications, the concept has not yet robustly emerged across the assembly industry. In light of its great potential, this paper presents the findings from a systematic literature review following the updated Preferred Reporting Items for Systematic Reviews (PRISMA) guidelines. It aims to provide an overview of the state-of-the-art learning from demonstration solutions developed for assembly-related tasks and offer a critical discussion of remaining obstacles in order to drive its progression towards meaningful deployments. The analysis includes a total of 61 papers over the period of 2013-2023 sourced from Scopus and Web of Science databases. Findings indicate that learning from demonstration has attained a significant level of maturity within the research environment, as evidenced by thorough experimental achievements, proving its great promise for industrial assembly applications. However, critical obstacles exist in the area of proven practicability, task complexity and diversity, generalisation, performance evaluation and integration concepts that require attention to promote its widespread adoption and create a seamless transition into industrial practices.

The escalating climate crisis requires innovative strategies across industries to mitigate environmental impacts, particularly in energy-intensive sectors like the process industry. This paper introduces the concept of Green Digital Twins as an approach towards achieving carbon neutrality in industrial energy systems. By leveraging Digital Twin technology, the study outlines a concept for reducing carbon dioxide equivalent emissions through improved energy allocations. The proposed functional components are the basis for further development towards sustainable practices. The research is grounded in the Design Science Research approach and draws upon extensive literature review and industry insights to establish requirements for Green Digital Twins. The proposed conceptual model focuses on functional components for data-based representation and is validated through application in a cyber-physical brewhouse as a representative of the process industry. This case study demon-strates the potential of Green Digital Twins to integrate energy data with individual processes, offering a novel pathway for industries to contribute to climate change mitigation. This approach provides a foundation for future innovations in Green Digital Twin technology and contributes to the ongoing discourse on digital transformation and sustainability.