The car body manufacturing process is an essential step in automotive production. It not only defines the structure and exterior of the car, but it must also meet all requirements of downstream processes. Therefore, one elementary goal is to ensure the geometric quality of the car body. The assembly of diverse free-formed metal components necessitates the utilization of a wide range of joining racks. Each of them is tailored to the distinctive form of the incoming components, the joining technologies employed and the prescribed joining sequence. Due to numerous influencing factors during the manufacturing process, the joining racks are designed to allow geometry adjustments at every contact point. Despite the extensive level of automation found in car body production, the process of achieving the desired geometry relies heavily on manual tasks. Since individual contact points within the joining rack are adjusted in iterative loops, the production line must be stopped every time. In an effort to improve the flexibility and automation level of the geometry processes within car body manufacturing, the mechanization of the adjustable contact points is imperative. To achieve this objective, it is necessary to first assess each contact point individually. This paper proposes a classification for contact points within joining racks to facilitate their assessment. After introducing five categories, a selection of different mechanization strategies is discussed for each one. Further research is needed to determine the applicability of the classification to a full-size automotive series production line.

In response to market shifts, manufacturing companies must innovate their production systems to sustain competitiveness. Incremental changes to production systems fail to address these challenges, necessitating radical innovations in manufacturing. This is particularly evident in the pharmaceutical manufacturing industry, where product shortages and an expanding product variety challenge the traditional mass production paradigm. However, there is a noticeable lack of methods and tools to support the initial stages of manufacturing innovation activities in the pharmaceutical and general manufacturing industries. This paper aims to fill this gap by drawing on insights from a radical manufacturing innovation project in a pharmaceutical manufacturing company. Based on the case study findings, six design principles for front-end radical manufacturing innovation processes and potential supporting tools and activities are proposed.

Model-based systems engineering is becoming increasingly widespread in the manufacturing industry. The reason for this lies in the potential for effective-ness, efficiency and complexity control offered by MBSE. In addition to the question of suitable tools, however, the design of appropriate product devel-opment processes for sustainably implementing MBSE poses particular chal-lenges for companies. In order to make the necessary adjustments to existing processes, companies need a uniform understanding of the elements to be adapted. This paper presents the findings from a systematic literature analysis of scientific work regarding product development and MBSE. As a result, ac-tivities, roles and models were identified as basic elements of the development processes to be adapted. A consolidated list of the identified elements provides an overview of the roles, activities and models for MBSE described in the lit-erature. The results provide users with orientation for process analysis and support the standardized description of processes in model-based product de-velopment. They provide the basis for further organizational design.

For the automotive industry, especially on the part of Tier 1 and Tier 2 suppliers, the future will be about maintaining sovereignty in the form of techno-logical openness and accelerating digitalization. The product portfolio, which is generally passed on from OEMs to suppliers for production, often includes body parts that cannot always be manufactured economically with the prevailing pro-duction technology. The reason for this is a high number of variants that require smaller batch sizes. Instead of producing such batches cost-intensively and un-profitably as part of large orders, they could be made attractive and profitable by producing geometrically individualized components with flexible variants and capacities. Scaling up variant-rich body sheet components for the economical production of small batches requires the introduction of new production technol-ogies. For this reason, highly flexible large-series production cells for car body sheet metal components that are scalable in all dimensions are being developed and tested. For the first time, they allow the process flow in series production to be redesigned on a component-specific basis. The aim is to reduce production costs for new, geometrically different component variants. On the other hand, a process generator develops the corresponding production plan. A digital repre-sentation of the manufacturing processes enables the selection of cost-, effi-ciency- and sustainability-optimized production chains depending on the number of parts. Established manufacturing processes for the production of car body components are supplemented in the cell by flexible designed processes.

The learning factory approach to technical competence development has proven valuable in full-time education, lifelong learning, and research over the past dec-ade. Learning factories for education and training have well-defined goals and consistent designs, focusing on enhancing participants' knowledge and experi-ence with specific technologies. Existing research on learning factories often pre-sents empirical cases but lacks systematic conclusions on how design aspects af-fect learning outcomes, hindering theory generation. This gap also means no standard frame of reference for defining a learning factory. The CIRP Collabora-tive Working Group's morphology is an initial attempt to characterise learning factories but understanding the design's impact on learning outcomes requires further study. Enke et al. addressed this by developing a maturity model to assess a learning factory's current state and identify areas for improvement. This paper investigates learning factory literature through an AI-supported review to identify trends in learning modes, target groups, topics, and didactics. A review of 421 studies found that most research focuses on transferring closed-case learning out-comes to students and industry professionals. However, there's a growing trend of knowledge creation within open-ended, interdisciplinary learning factories. These insights help conceptualise Learning Factories as a multimode concept, requiring distinct design principles based on the learning target, whether for knowledge transfer or creation, and supporting strategies for cost-reduction, knowledge extension, or innovation.

Global supply chains account for 80\% of international trade but pose significant human rights and environmental risks. The EU Corporate Sustainability Due Diligence Directive and similar regulations impose new compliance requirements, challenging companies to integrate sustainability, due diligence, and supplier relationship management. This paper reviews the operational challenges of complying with these regulations, including supply chain dilemmas related to regulations, complexity, and transparency. It explores how risk-based and collaborative approaches and leveraging technologies can promote effective sustainability due diligence in global supply chains. This study contributes to understanding how large firms can navigate evolving supply chain regulations to ensure compliant, sustainable practices by ad-dressing challenges and mitigating strategies.

Despite the growing conceptual emphasis on Intertwined Supply Networks (ISNs), there is still a lack of simulation-based studies that quantitatively assess their operational adaptability under disruptions and uncertainty. This study develops a discrete-event dynamic simulation model using Py-thon for an intertwined pharmaceutical-food supply network, to evaluate the effectiveness of adaptive sourcing and dynamic re-routing under three scenarios: baseline, supplier disruption, and transportation disruption. The model tracks operational Key Performance Indicators (KPIs) such as back-log, service level, and transportation cost. Comparative analysis of all three scenarios show that adaptive mechanisms reduce backlog by 61\% and transportation costs by 14.2\%, demonstrating substantial improvements in resilience and efficiency. The approach offers valuable insights for design-ing more resilient and adaptable supply networks under uncertainty.

Remanufacturing is widely regarded as an end-of-life strategy, that can improve sustainability and resilience. Remanufacturing restores components to their original functionality, potentially allowing them to reintegrate into standard inventory procedures alongside non-remanufactured components. By going through the literature on design for remanufacturing, this study identifies critical parameters for assessing the ability to remanufacture components, including the ability to disassemble, clean, inspect, repair, upgrade, and test, as well as the cost/resource efficiency of these processes. Based on these parameters, a decision support model is proposed to evaluate whether the components are suitable for remanufacturing. An illustrative example of a circulating pump demonstrates the practicality of the model, distinguishing components suitable for remanufacturing from those better suited for recycling. The findings highlight the potential of integrating a decision support model to identify components suitable for remanufacturing, and how such a model can enhance sustainability and resilience.

In recent years, research on supply chain disruptions and internal failures has in-creased due to their significant impact on extended downtime, reduced revenue, and diminished customer trust. Traditional reliability and risk management approaches primarily focus on failure prevention, whereas resilience extends beyond by incorporating prediction and adaptive strategies. Current methods for measuring resilience rely on classical reliability metrics, such as mean time be-tween failures (MTBF) and mean time to repair (MTTR). While useful, these metrics fail to capture the multi-dimensional aspects of resilience, specifically the ability to prevent, predict, and adaptively respond to breakdowns. This limitation leaves production systems exposed to critical and unanticipated failures, making it difficult for practitioners to prioritize investments in predictive maintenance technologies or design effective resilience-enhancing strategies. To address this gap, this study develops a quantitative, data-driven resilience assessment framework tailored for production machinery. The framework employs Bayesian Networks (BNs) to model probabilistic dependencies among machine components, failure modes, and resilience strategies, enabling a holistic evaluation of machine resilience performance. A case study on a compound forming machine demonstrates the practical application of the methodology, highlighting its ability to identify critical failure modes and support decision-making for fault detection and human-in-the-loop assistance systems. By facilitating data-driven decisions and targeted retrofitting with emerging technologies, this approach enhances production resilience, reduces downtime, and provides valuable insights for both industrial practitioners and researchers.

Traditionally, in industrial production systems, the reconfiguration of machinery relies on the expertise of human operators. This expertise is accumulated over time as the worker gains experience. The actions and intentions of experienced operators can be recorded with motion capture technologies such as body and eye-gaze trackers. This data can potentially support inexperienced or less proficient workers in acquiring knowledge to perform similar tasks by sim-ulating the actions of the experts. This work presents a method to capture and analyze the motions of an expert operator while reconfiguring machine tools of the forming machine. The findings reveal that actions associated with reconfiguring machine tools for forming press machine can be recognized and segmented by applying time-domain analysis on the motion data. The results show potential of developing a knowledge-based assistance system that encapsulates the expertise of operators and supports novice operators to replicate learned actions effectively in order to enhance productivity on the shop floor.

As manufacturing systems have become more complex, uncertainties and disruptions have increased, where resilience has become essential for manufacturers, in order to ensure production continuity, and quick adaptations to disruptions such as equipment failures, supply chain issues, or the introduction of new products. This work presents an approach of assessing resilience in steel manufacturing, using the Penalty of Change (PoC) methodology. Combining PoC with a heuristic scheduling algorithm, the goal is to simulate and evaluate the impact of disruption. Resilience is quantified by measuring the degree of production deviation caused by disruptions such as machinery breakdowns and unplanned orders with high priority. Through a case study of a steel manufacturing plant, the proposed framework offers a clear methodology for assessing resilience and identifying areas for improvement. This study emphasizes the importance of resilience in manufacturing, offering a practical tool for decision-makers to optimize production planning and enhance operational robustness under certain conditions.

Resilience in production environments is defined as the system’s ability to avoid, adapt, and rapidly recover from disruptive events, attaining normal operational conditions with minimal time and financial investment. Current methods lack the capability to quantify risks based on real-time data from production devices, creating a critical gap. This article presents a novel method aimed at enhancing resilience by integrating data from micro and meso levels within precision machining production plants. By utilizing real-time monitoring data, from machines and production processes, the proposed approach is based on the assessment of risks of failure of machines and early detection of anomalies in machining processes, to support decision-making. Leveraging the Industrial Internet of Things (IIoT), the method collects and analyses real-time machines and machining process data for risk assessment and early anomaly detection. The proposed method minimizes the impact of unexpected failures by incorporating failure risk calculations into plant-level decision-making. Validation of the method was conducted using a milling machine in a controlled laboratory setting, demonstrating the significance of real-time data and health assessment in developing resilient industrial systems. Preliminary results indicate strong potential for industrial application, with the next step being implemented in a production plant.

Resilience is increasingly critical in modern manufacturing, especially within complex supply chains. Information sharing based on data sovereignty prin-ciples enables informed, timely decisions, improving supply chain resilience. Data space technologies, like International Data Spaces (IDS) and Gaia-X, support secure, multi-stakeholder data sharing. This work integrates IDS and Gaia-X data spaces into a data spaces framework for resilience (DSF4R). The framework facilitates secure data exchange using a compute on demand environment, ensuring data remains secure within the provider's system while enabling external services to process it. This eliminates the need for data transfer, ensuring privacy and compliance. Applied in a real-world in-dustrial use case from the steel industry sector, it demonstrates enhanced re-silience and broader applicability.

Reconfiguration of machine tooling components in industrial production systems is highly dependent on the expertise of human operators. This expertise is developed over time through the experience of skilled workers. A sig-nificant challenge is the absence of a predefined set of actions and their sequences. Moreover, due to variability in user behavior, structured tasks are often performed with diverse patterns and sequences that differ across users and con-texts. This work analyzes the variability in reconfiguration patterns caused by differences in action sequences of an operator, by employing an action prediction model that annotates action and sequence semantically. The findings are evaluated based on experts’ action duration and sequence. Results highlight the potential to identify skillsets for reconfiguring machine tooling in forming processes use-case, paving a way for systems to enhance their productivity on the shopfloor.

This paper introduces Rebound Reasoning, an architecture leveraging itera-tive prompting and mixture of agents to improve the reasoning capabilities of quantized language models for efficient deployment as services in pro-duction systems. Our approach enables quantized models to narrow the per-formance gap with unquantized counterparts while requiring only a fraction of computational resources, making them suitable for manufacturing envi-ronments. We demonstrate the effectiveness of Rebound Reasoning through comprehensive experiments using the Alpaca Eval benchmark. The architecture integrates with vLLM and ollama infrastructure, generat-ing multiple perspectives through iterations to reduce biases and synthesize more comprehensive answers. This approach enhances output diversity and quality while improving error tolerance and consistency—particularly val-uable for decision support in production environments. Our experiments show Rebound Reasoning achieves comparable perfor-mance to unquantized models while offering significant improvements with quantized models. The approach's flexibility allows adaptation to various model configurations, making it scalable for different production applica-tions. This work contributes to making large language models more efficient and accessible as services for production systems, broadening the applica-tion of AI reasoning capabilities in manufacturing environments.

In recent years, large language models (LLMs) have gained significant popularity, driving innovation across various domains. LLMs are widely used across fields, including general-purpose robotics. However, their use in industrial applications still has a lot of potential. This paper presents a prototypical implementation of a collaborative robot system which uses a set of carefully instructed LLMs to plan a collaborative assembly/disassembly process. Using natural language descriptions, the system generates action plans, assigning tasks to robots or humans based on capabilities and tools. Central to this implementation are state-of-the-art object detection and a reasoning process to deduce action plans. In addition, we implemented a monitoring system employing sequential LLMs to check for and correct errors during task execution. We detail the integration of system components, including peripherals and prompt engineering methods to optimize task outputs. To validate the approach, we demonstrate its application on the assembly and disassembly of a Raspberry Pi-based product. This work explores how LLMs enhance human-robot collaboration (HRC) in industrial assembly and disassembly, improving task planning, real-time error detection, and adaptive correction.

With increasing product complexity and shorter product lifecycles, manufactur-ing companies – especially small and medium-sized enterprises (SMEs) – need to adapt their production planning processes quickly and reliably. A key step is the creation of accurate and easily interpretable work instructions for manual assembly tasks. Using Controlled Natural Language (CNL) for these instruc-tions offers distinct advantages: it preserves the readability of natural language while enforcing a standardized syntax and vocabulary. In this context, recent advances in Large Language Models (LLMs) provide an opportunity to stream-line the generation of CNL-based assembly instructions. This paper presents a proof of concept that investigates whether an LLM, coupled with a Retrieval-Augmented Generation (RAG) approach, can generate CNL-compliant instructions from Product-Process-Resource (PPR) models. Two representative use cases demonstrate the potential of this approach to im-prove structural consistency and reduce manual effort. However, the evaluation remains limited to a small number of examples and a single LLM, and current PPR models lack certain critical information such as precise positioning. These findings indicate that while LLMs show promise for supporting automated gen-eration of CNL-based instructions, further work is needed to refine domain ad-aptation, improve data quality, and validate the approach with broader datasets and alternative language models. This work thus lays the groundwork for more flexible, automated production planning processes that exploit the synergy of LLMs and controlled linguistic frameworks.

Human motion modelling is essential in understanding and enhancing human-to-human collaboration, particularly during the collaborative handling of rigid objects. This paper presents a data-driven approach to create a new model for human motion in collaborative object-handling scenarios to improve the effectiveness and adaptability of collaborative tasks. This study also addresses the accurate analysis of captured data from human-to-human interactions. Through an experimental setup utilizing a motion capture system, human motion data is collected and preprocessed for a collaborative handling task to create a model that can generate motion styles. The proposed model is trained based on experimental data by using Gaussian Mixture Models (GMM) and functional Principal Component Analysis (FPCA). The performance of the proposed model is evaluated for its spatial accuracy. The results demonstrate that the model is capable to capture the dynamics of human-to-human interaction and motion styles, which makes it feasible for the collaborative process and potential applications in human-robot interaction (HRC).

The automotive industry faces major challenges due to shifting market demands and technological advances, particularly in balancing innovation, cost efficiency, and growing product variability. Flexible manufacturing techniques (FMTs) with automation support adaptability to diverse component designs. Despite progress in 3D modeling, 2D technical drawings remain essential due to their compatibility with existing workflows. Automating their processing is therefore crucial for their seamless integration into modern manufacturing systems. This paper presents an AI-based approach for analyzing 2D technical drawings by detecting geometric features (e.g., reinforcements, embossments) and extracting manufacturing-relevant text (e.g., dimensions, tolerances). The prototype uses Faster R-CNN for object detection and a Keras-OCR-based pipeline for text recognition. A dataset of 204 labeled drawings covering eight feature types was created. Results confirm the system’s ability to identify complex patterns and extract critical data for downstream automation. The approach provides a scalable solution, especially for legacy components lacking CAD data and relying on scanned 2D documents.

The special machinery sector is distinguished by a high level of complexity, driven by the consistently growing demand for customer-specific special solutions. In today's volatile market and highly competitive landscape, suppliers must be able to respond swiftly with tailored offers for specialized machinery. It is therefore essential to adopt a success-oriented approach to offer management, ensuring the detailed preparation of offers. One potential strategy for addressing these challenges in the design process is to prioritize the reuse of individual units (artifacts) from previous projects in new customer projects. Information to support the design process and the calculative preparation of an offer is relevant here. This article provides an overview of reuse strategies for the design of production systems. The approaches and mechanisms of reuse are described in more detail. We provide an overview of current research gaps in this field and a derivation of possible further developments for the development of a supplementary methodology for use in offer management in mechanical engineering.

Using a balanced two-wheeled scooter as an example, the development of an automated design process for a hybrid assembly system is illustrated. A framework of graph-based design languages is used for the modeling, starting with an ontology which defines the concepts and their relationships in Unified Modeling Language (UML) syntax. Based on this, a design graph is generated using a rule-based approach. Furthermore, the question of how manual assembly steps can be simulated and analysed as part of an automated design process is addressed. All processes are automatically derived on the basis of the product model and the associated joining information, which form the basis for a 2D-and 3D-layout of an assembly cell. The manual assembly system is evaluated via an Unreal Engine 5 interface using the created layout. The virtual execution of the assembly enables the recording of movement data, which serves as the basis for an economic simulation. The results of this simulation provide the foundation for continuous adaptation of the assembly system, ensuring that the assembly process is designed in accordance with economic principles. The architectural design allows for the validation of hybrid assembly systems in their developmental phase.

As companies increasingly adopt service-based business models, the design of coherent and scalable service architectures becomes essential. While literature highlights the need for service analyses as a precursor to architectural design, it lacks concrete methodologies that integrate data from various life-cycle stages and organizational domains. This study addresses this gap by proposing a structured analysis method for identifying hierarchical decompositions across digital, physical, and mental sources—spanning sales, invoicing, and operations. The research question guiding this work is: how can the scope of service architecture analyses be systematically expanded to include more preexisting system decompositions, in order to reduce knowledge-based implementation barriers? The proposed method is tested in a case study at a Danish third-party logistics (3PL) company, revealing structural and semantic misalignments across organizational systems. By increasing transparency and highlighting misaligned service structures, the analysis improves the foundation for conceptualizing and implementing service architectures.

Manufacturing companies have a strong impact on climate change due to their immense degree of greenhouse gas emissions. Hence, it is mandatory that these companies take measures to reduce their greenhouse gas emissions. A promising measure is the reduction of the carbon footprint caused by single processes within manufacturing process networks. Various approaches to reduce the carbon footprint of single manufacturing processes have already been described in literature. However, there is a scientific need for action when it comes to the systematic reduction of the carbon footprint of queueing systems which are typically an integral element of manufacturing process networks. Therefore, this paper presents a novel approach that aims to assess and reduce the carbon footprint of manufacturing queueing systems systematically. First, a general queueing system is defined and its variables that contribute to the carbon footprint in manufacturing are determined. Subsequently, a mathematical model is designed which enables an abstract description of the interactions of different input variables on the carbon footprint of queueing systems in manufacturing. Afterwards, recommendations to reduce the carbon footprint of queueing systems are developed based on the previous findings. Finally, the approach is validated for a specific manufacturing queueing system with a discrete-event simulation study. The results give foundations for further research on dependencies within manufacturing process networks.

The aerospace industry’s drive to improve fuel efficiency and reduce emissions has led to the widespread adoption of lightweight, thin-walled structures in aircraft assemblies. However, these flexible structures are susceptible to deformation under gravitational and external forces, causing assembly deviations and potential misalignment. To counteract the misalignment caused by part deformation in joining processes, shims are specifically designed to fill the gaps between components. Yet, despite advances in metrology, the process of shimming remains largely manual, since each assembly’s unique gap profile requires physical measurement, fitting, and adjustment, which hinders full automation. A promising solution is the use of a digital twin to simulate and predict part deformation during the joining process. The predicted deformation resulting from the digitalization of the compliant structures to be joined can be used for designing shims in a virtual assembly environment or as input for adaptive tooling systems specifically designed to compensate for the deformation. This work introduces a concept for a test cell equipped with a large-scale metrology system to capture the data needed for constructing a digital twin of manufactured large, compliant, thin-walled structures. Corresponding data pipelines, designed to meet FAIR data principles, combine a time-series database for continuous measurement data with a non-relational database for storing measurement metadata. Preliminary results demonstrate the test cell’s ability to accurately capture deformation data in real-world conditions. These findings lay the groundwork for future digital twin development and offer insights to improve alignment precision in the assembly of large, compliant, thin-walled structures.

Innovation Systems are dynamic ecosystems that integrate individuals, organizations, policies, and processes to drive innovation, economic growth, and employment. Building on the Co-Innovation Cosmos framework introduced in a previous paper, this work focuses on two core elements: collaborative upskilling and customized connectivity. The concept of Focus Parcours is introduced to facilitate specialized, adaptable learning pathways tailored to regional, industry-specific, or thematic needs (e.g., Leadership or Intrapreneurship Parcours). These pathways strengthen Local, Regional and National Innovation Systems by enhancing targeted education and skills development. Connectivity emerges as a cornerstone of innovation ecosystems, blending physical infrastructures with digital layers to foster seamless integration. Human-centricity remains at the core of this approach, with the Co-Innovation Sphere - a novel digital overlay - connecting the proposed innovation infrastructure and enabling cross-border knowledge exchange within a Global Innovation System. This paper highlights customizable elements and the flexibility of the proposed strategy, enabling a holistic strategy to infrastructure usage, tailored upskilling efforts and connectivity solutions to specific goals - whether enhancing regional innovation capacities, fostering industry-specific collaboration or addressing global challenges. By aligning education, infrastructure and collaboration, the approach aims to cultivate a transnational innovation mindset and promotes sustainable economic growth.

In recent years, companies have been struggling to best cope with the in-creasing request for customized and personalized products that characterizes mass customization paradigm. The efficient management of product variety requires effective solutions to make companies able to deliver such products in a short time and at a high-quality rate. The Delayed Product Differentiation (DPD) is rising as one of the most effective strategies to manage mass customization, implemented in practice using product platforms. Platforms are intermediate products formed by the most common components within a product family and managed through a make-to-stock strategy. After the arrival of the customer order, platforms are transformed into the final variants through assembly/disassembly customization tasks, delaying the final prod-uct differentiation point. In such a scenario, this paper proposes an innova-tive decisional algorithm for product platform design and selection to reduce the overall effort for platform customization into final variants. The applica-tion of the procedure to a reference industrial case study showcases its rele-vance and validity in managing high product variety.

Uncertainty is an enduring facet of our world, shaping individual experiences and societal dynamics, including business processes and strategies. While challenging, uncertainty also serves to catalyze creativity, a foundational concept of mass customization (MC) co-design toolkits (i.e., product configurators). The present research explores whether the MC experience could be deliberately designed to foster uncertainty to nudge a variety of consumer types to embrace unknowns. This promotion of uncertainty may seem counterintuitive due to uncertainty’s role in fostering complexity which acts as a significant barrier to delivering value-laden MC experiences. However, the interaction of uncertainty and complexity presents an opportunity to enhance consumers’ creative engagement. The MC co-design toolkit, therefore, holds a potential role in empowering individuals struggling with the ambiguity of uncertain outcomes through cultivating confidence during any challenges of the MC experience. This approach envisions a toolkit to facilitate collaboration between consumers who thrive in uncertain environments and those whose risk aversion inhibits their inherent creative potential. This interactivity can yield intangible relational benefits, including self-realization and enhanced individual and societal well-being. Moreover, optimizing the MC toolkit to embrace creative uncertainty could offer providers unique observational insights into managing the unpredictability inherent in consumer-driven creativity. By leveraging concepts of Knightian uncertainty and behavioral economics, particularly ecological and bounded rationality, the present research evaluates the transformative potential of designing MC experiences to harmonize creativity and uncertainty. Keywords: Mass Customization, Co-Design Toolkit, Product Configurator, Creativity, Thinking Style, Knightian Uncertainty, Bounded Rationality, Ecological Rationality, Behavioral Economics

In observing the evolution of product configurators since 2007 in our Configurator Database Research Project we have been conducting analyses on the development of configurators based on various criteria as well as exploring the appearance of their user interfaces. Being a crucial tool in the concept of mass customization the criteria for successful product configurators have changed and adapted to the evolving needs of customers and the technological possibilities. The rise of Artificial Intelligence (AI) is and will massively influence the way how software interaction is taking place. In our paper we examine if AI will augment and enhance a product configurator experience and what features and effects might change in how configurator interaction will take place in the future

Efficient production planning, machine optimization, and bottleneck analysis are critical for reducing cycle times in manufacturing processes. This paper introduces an automated process mining approach designed to streamline these tasks by generating and optimizing finite state machines (FSMs) from event logs derived from signal changes in industrial plants. Unlike conventional methods, this approach requires no prior knowledge of the configuration or behavior of the target programmable logic controllers (PLCs). The synthesized FSMs facilitate the extraction of process-oriented insights, enabling the identification of underperforming manufacturing processes and detailed analysis of their durations. The proposed method is validated through a case study on a real industrial plant, demonstrating its efficacy in uncovering process inefficiencies and supporting decision-making. This work provides a novel, generalizable framework for process analysis in manufacturing environments, contributing to the broader field of automated process optimization.

Process time variance is frequently assessed in assembly line balancing research, a crucial production planning step. However, it has been largely overlooked in planning of highly automated car body construction assembly lines, owing to the deterministic nature of robots. Resistance spot welding, the most revalent joining technique, employs feedback control systems for quality assurance, leading to varying process times. When multiple welds are done per cycle, these variations lead to fluctuations in overall cycle times. A quantitative analysis of 16 assembly lines reveals a 0.6s median cycle time range. By incorporating process time variance in the planning of car body construction, this cycle time variance can be controlled to mitigate effects on assembly line performance. However, modelling of process times is required because these cannot be known in advance. As simulations on the balancing of car body construction assembly lines indicate, the model accuracy, however, needs to be high. Inaccurate distributional shape and spread predictions cause deficient planning assumptions, resulting in unpredictable outcomes.

In Industry 4.0 and 5.0 eras, Reconfigurable Manufacturing Systems (RMSs) rose as effective solution to meet the highly dynamic customers’ needs. RMSs deal with intelligent machines able to perform a large variety of tasks, by ‘reshaping’ themselves thanks to auxiliary modules which can be assem-bled and disassembled to/from their base structures. However, despite auxil-iary modules increase reconfigurability level of these systems, their frequent assembly/disassembly and the highly different working parameters and con-ditions subject them to greater wear and tear risks. Hence, an appropriate maintenance policy must be planned to maximise the modules’ lifespan, re-ducing downtimes and costs. In this context, embracing the progress of digi-tal technologies, Digital Twins (DTs) are becoming a widespread tool in smart manufacturing for performance analysis, process simulation and the development of what-if DfX analysis. This paper presents a dynamic DT modelling environment, based on a commercial 3D scale simulation plat-form, to integrate maintenance policy in RMS operations’ management. By focusing on the most relevant challenges posed by RMSs, the model allows simulating the impact of responsive and flexible maintenance alternatives, tracking their impact on the system uptime and productivity. The proposed DT modelling environment is preliminary tested through a simplified indus-trial case study inspired from the metalwork mechanical sector, showcasing its effectiveness in supporting operations’ activity planning and management within complex and adaptable production environments.

Automobile manufacturers face increasing final assembly complexity due to expanding model portfolios, customized configurations, and emerging powertrain technologies. Conventional assembly lines struggle to absorb process variability efficiently. In response, this study introduces a hybrid assembly concept that integrates cycle-based line segments with flexible matrix segments to combine efficiency with adaptability. A structured scoring-based decision methodology is developed to allocate process modules based on criteria such as process time variance, variant diversity, labor requirements, and precedence constraints. The approach is validated through a case study at a German automotive OEM. Results show that 11 out of 17 assembly clusters benefit from matrix integration, with reduced cycle time losses and improved flexibility. The findings demonstrate the practical applicability of the methodology and its potential to support the transition toward future-oriented, modular assembly systems.

This paper introduces a novel approach for improving the adaptability and scalability of modular assembly systems by integrating a decentralized, agent-based control with state-based planning. Traditional assembly systems struggle with adapting to varying production demands due to their reliance on fixed, centralized workflows. Our proposed concept leverages state-based descriptions of assembly processes and a multi-agent system to enable real-time reconfiguration and dynamic process planning. This approach allows seamless adaptation to changes in production volume and product types, reduces installation times, and extends the operational life of assembly equipment. The results indicate potential improvements in flexibility and resilience compared to conventional methods, suggesting promising directions for future research and applications in manufacturing environments, especially those requiring increased adaptability to market changes.

The circular economy offers a potential solution to a multitude of contempo-rary challenges, including those pertaining to resource consumption, import dependency, and economic growth. This is accomplished by facilitating the circular reuse of products through the implementation of various R-strategies. Of these, remanufacturing is a particularly promising approach, as it aims to preserve as much of the value-creating activities that were previ-ously invested in the product as possible. This is founded upon the disas-sembly of end-of-life (EoL) products, which may exhibit considerable varia-tion in geometric dimensions, arrangement of parts, and materials. The high variance of EoL products, especially evident in WEEE (Waste Electrical and Electronic Equipment) products significantly complicates automation and imposes extremely high demands on the flexibility of gripper systems. This challenge currently hinders the economic operation of automated disassem-bly systems, as complex and highly adaptable gripper systems are required, which are often customized for single-product applications. Accordingly, this paper proposes a conceptual framework for a flexible and reconfigurable gripping system for robotic disassembly. This system is capable of automati-cally reconfiguring to process a variety of products. To develop this system, a functional analysis is conducted, in which various gripping systems are ana-lyzed to identify requirements and functions of adaptable handling systems for the disassembly of WEEE components. This is an important step toward developing more agile, flexible, and cost-effective disassembly systems.

Reconfigurable Manufacturing Systems (RMS) offer a compelling solution to anticipating uncertainty in volatile markets. Despite their potential value, their adoption is often constrained by industrial investment models that narrowly focus on short-term business cases, overlooking lifecycle considerations and the economic value of flexibility. This oversight can be traced to inadequate systematic risk management of flexibility dimensions and insufficient planning for extended equipment use in production system design. This paper explores how companies can better anticipate market uncertainty in their economic investment models, a prerequisite for valuing reconfigurability effectively. This study identifies critical barriers to adopting RMS through a literature review and interviews with representatives from four manufacturing companies. Findings reveal a gap in including and quantifying the economic and sustainability advantages, particularly regarding lifecycle and flexibility considerations caused by volume, product, and product mix uncertainties. Instead, current investment models include short-term cost metrics, leaving long-term adaptability undervalued.

The starting point of this study is the increasing number of variants with simultaneously decreasing batch sizes of body sheet metal components. The suppliers' current production and plant technologies are increasingly reaching their economic limits. The SkaLaB joint project aims to provide economically viable solutions through scalable and flexible production processes and strate-gies. Car body sheet metal components have small-scale, locally distinctive yet characteristic geometries that realize the functional properties of the component. In production, secondary forming elements represent critical areas whose manu-facture poses particular challenges and limits the flexibility and universality of the tools. A flexible and scalable tool system for the universal forming of sheet metal components is being developed for the separate production of secondary forming elements, independently of the main forming process. Within the scope of this work, individual groups of secondary forming elements are classified, and an order of these elements is created. In addition, a tool set for two particularly relevant secondary forming elements is designed and tested experimentally. Fur-thermore, a spring-loaded clamping system for rubber pad forming is developed, which enables a localized application of this process within a larger sheet metal blank.

The increasing need for flexibility in industrial automation has prompted the de-velopment of adaptable grippers capable of handling objects with varying shapes and sizes. This paper introduces a novel pneumatically actuated multi-fold gripper with a programmable deformation mechanism. The core of the gripper is a flexi-ble ring actuator with an internal air channel. By inserting star-shaped forms dur-ing the first actuation, folds formation in the ring’s inner wall can be controlled, allowing for the definition of discrete folds. Experimental tests using inserts with 4, 5, and 8 tips demonstrated successful and repeatable programming of the ring’s deformation pattern, which remained stable over multiple actuations and af-ter a one-week rest period. Attempts to generate only two folds were unsuccess-ful, likely due to uncontrolled stress distribution and material limitations. The multi-fold gripper is manufactured additively using a combination of SLS and PolyJet technologies, enabling rapid customization. This method provides a sim-ple, low-cost approach to pre-shaping soft grippers, reducing the energy losses typically associated with deformation. The concept of using localized structural memory introduces a new design pathway for soft robotic actuators. Future work should explore automatic deformation programming through embedded structural features created during the additive manufacturing process.

Research in the field of automated disassembly, in relation to end-of-life electric car batteries, has focused on the removal of screws over the last ten years. The most common detection system is based on visual recognition, which is unable to recognize the pitch or length of the screw. The unscrewing process therefore relies on knowledge from databases, human cooperation, or force sensors to regulate the movement of the robot in the direction of the screw axis during unscrewing. This movement is crucial during the screwing process to maintain the screwdriver's engagement with the screw drive and ensure the successful removal of the screw. In this paper, a new cost-efficient solution is presented in which a spring system and a foil potentiometer are embedded in a conventional industrial fixed screwdriver to measure the spring travel caused by contact between tool tip and screw head. Tests were carried out to determine the accuracy and precision of the spring travel measurement. The control of the robot's movements via flexible G-code programming is explained and mathematically proven to be safe, depending on the robot's feed rate and its processing time. Finally, practical experiments were performed to verify the correct execution of the process. The presented technology is intended to be easily reproduced and integrated into any CNC-controlled robot equipped with a conventional screwdriver to make the disassembly technology more accessible.

Interpreting human motion during physical human-robot interaction is critical for enhancing robot adaptability and acceptance in collaborative tasks such as joint handling of rigid objects. Effective modeling of the trajectories of human joints from initial to target positions allows robots to adapt their motion accordingly. However, the dynamics of role shifts between human and robot during such adaptation remain insufficiently understood. This work presents a trajectory-based approach for modeling role transitions in human-robot collaboration (HRC). This method analyzes situations that trigger role-switching and interaction conflicts by evaluating force and velocity vectors at the end-effector level. A finite state machine (FSM) is implemented to classify interaction states, such as following, leading, or resisting, based on spatial deviation and motion alignment. In addition, this method monitors force unit within a desired threshold while planning robot motion along a generated human motion corridor. Unlike traditional pre-programmed robotic systems, this role-adaptive approach enables flexible and responsive behavior, supporting more natural and robust collaboration in dynamic production environments.

The integration of collaborative human-robot systems (HRCS) into manufacturing has revolutionized industries, enabling flexible, efficient, and human-centric automation. However, safety validation remains a significant bottleneck, with traditional methods often manual, costly, and time intensive. To address this, we propose the Scenario-based Collaborative Robot Evaluation (SCORE), a novel toolchain leveraging scenario-based simulations, automated risk assessment, and AI-driven adaptability to streamline safety validation for HRCS. SCORE focuses on developing a descriptive language to comprehensively model human-robot, robot-robot, and robot-machine interactions, including hardware setups, sensor configurations, and spatial distributions on the shop floor. This descriptive language serves as the foundation for generating a range of simulation trials, simulating dynamic interactions between humans, robots, and machines under diverse operational scenarios. Unlike traditional methods, SCORE allows users to input setup description, safety and/or control algorithms, enabling automated validation against specified standards and requirements. This process identifies risks and evaluates system performance, providing a robust framework for iterative improvement. By leveraging advanced simulation platforms, such as NVIDIA Omniverse and ROS2, and integrating user-defined safety measures, SCORE offers a flexible, scalable, and cost-effective solution for validating HRCS setups. This conceptual paper outlines SCORE’s methodology highlights its innovative use of descriptive modeling, and discusses its potential impact on enhancing safety, efficiency, and adaptability in collaborative robotics.

This study introduces a simplified approach to model the energy consumption of a collaborative robot that strikes a balance between accuracy and ease of integration with existing standard planning tools. By combining differentiable inertial and kinematic models with an adaptable single-parameter electrical model, we can significantly reduce the complexity of the identification process. The validation results show that both proposed models closely approximate actual energy usage, exhibiting minimal deviation and ensuring their suitability for practical applications. In future research, we aim to expand our modeling approach to industrial robots and validate its effectiveness across a diverse range of robotic systems.

Remanufacturing transforms End-of-Life (EOL) products into high-quality, sustainable alternatives, contributing to significant economic and environmental benefits. However, due to the unpredictable state of EOL products, remanufacturing remains a complex and uncertain process. Human-Robot Collaboration (HRC) offers a human-centric, flexible approach to address these challenges while enhancing operator’s overall well-being. This study proposes a methodology for designing and evaluating collaborative remanufacturing workstations by identify-ing optimal HRC scenarios for different tasks. Various production modes are as-sessed based on process quality, efficiency, and sustainability. A novel evaluation framework incorporating a Quality Index (QI) and a Sustainability Efficiency In-dex (SEI) is introduced to compare alternative workstation configurations. The approach is validated through a case study inspired by the household appliances industry, specifically focusing on domestic refrigerator disassembly for remanufacturing. The results highlight the superiority of HRC-based setups, demonstrating advantages over manual and fully automated processes in terms of quality, flexibility, ergonomics, and sustainability.

In soft robotics, pneumatic network actuators (pneunets) enhance safety in human-robot collaboration by reducing injury risks from rigid grippers. This study investigates the additive manufacturability and bending behavior of pneunets that are fabricated using various additive manufacturing (AM) processes, focusing on the Liquid Additive Manufacturing (LAM) process for real silicone. Compared to TPU-based pneunets manufactured via Fused Filament Fabrication (FFF) and Selective Laser Sintering (SLS), LAM pneunets achieve a larger bending angle (104°) without structural failure and exhibit superior resilience compared to PolyJet (PJ) pneunets. Trajectory analysis reveals that softer materials experience pre-bending due to gravity. However, the LAM pneunets demonstrate high elasticity and return to their original shape after actuation, making them promising for soft robotic applications such as delicate object handling. The findings highlight the potential of LAM manufactured silicone to improve pneunet performance. Future work focuses on optimizing the LAM process, evaluating long-term durability, and refining trajectory control to enhance functionality, for more flexible and efficient soft robotic systems in industrial settings.

Fiber composite materials balance strength and mass criteria. Current practic-es rely on multi-stage manual operations for hand layup solutions. Alterna-tive methods need to be developed to enable effective automation to be in-troduced in industrial domains. Specialty flexible grippers have been de-signed for dry, limp material gripping leveraging origami principles, and in this research, the proposed robotics systems integrated is presented. Design guidelines are presented along with solutions for gripping and wrinkle elimi-nation. Modular experimental validation activities were performed for the grippers, different mold surfaces, and a prototype bladder based strategy for smoothing out the positioned material. The findings of the research pave the way for rapid, efficient, mechanically simple, and cost-effective, integration of origami based compliant mechanisms in fabric composite manufacturing settings.

Cranioplasty implants are applied for medical cure of human cranial diseases, e.g. skull cancer, malformation or cranial trauma after accident. Particular crucial for long term curing is the acceptance of the implant by the human body and its long-term integrity, which is generally achieved using biocompatible materials such as grade 5 titanium. As each medical implant is an individual product, flexible and adaptive manufacturing processes are in need to allow for an economic and auto-mated fabrication. In the past, metallic implants have been fabricated directly by subtractive or additive processes. In an experimental approach, these implants were fabricated by forming of sheet metal plates in 3D-printed tools from poly-mer. In the light of this feasibility study, a rapid tooling process from cheap, re-cyclable materials, which can be applicated to directly manufacture molds from spatial scan data is demonstrated. Based on sheet metal forming, the presented process minimizes material effort of highly expensive titanium alloys while at the same time allows for the economical and rapid fabrication of individual single plate implant parts.

Additive Manufacturing (AM) processes enable the validation of design vari-ants, and the manufacturing of low volume specialty components. A produc-tion on-demand solution can be established close to a customer, but the economies of scale need to be considered. Slow fabrication times are an is-sue for larger production volumes, but for the directed energy deposition (DED) and hybrid manufacturing (where additive and machining operations are interwoven), new process planning scenarios can be explored for both low and medium volume production levels, which aligns well with address-ing on-demand service and out of production components. Prior to exploring multi-function machines and dynamic layouts, the precedence diagrams need to be determined as well as the process summary matrices. However, with DED and hybrid AM, there are unique scenarios for the dependencies and thermal cycling conditions. This preliminary research focuses on defining a framework for DED AM precedence diagrams, process summary data, and insights for systematically decomposing components for macro level process planning.

Considering increasing demands for adaptable manufacturing systems, soft pneumatic grippers – particularly those based on pneumatic networks (pneunets) – offer promising potential due to their intrinsic compliance and compatibility with additive manufacturing. However, their limited industrial adoption is due to concerns regarding long-term durability, which remains underexplored. This study investigates the long-term bending behavior of 3D-printed TPU pneunets with varying wall thicknesses. Using cyclic bending, the deformation trajectories of the pneunet tips are tracked and analyzed. Results reveal a consistent trajectory shift during initial cycles, which diminishes over time and can be effectively de-scribed using a logarithmic function. The model’s accuracy is validated through high coefficients of determination (R² > 0.95), and the pressure-dependent change is further approximated via a quadratic fit. Findings indicate that wall thickness has no significant impact on long-term deformation. This work under-scores the need for standardized testing protocols to enable reliable cross-study comparisons and support the broader application of soft robotic grippers in in-dustrial contexts.

With the development of electromobility, car manufacturers have established and expanded the value chain for lithium-ion batteries. Specialised logistics for battery systems can help improve the environmental performance of the supply chain by increasing its efficiency and sustainability, taking into ac-count the requirements for volumes and characteristics of battery systems and cell modules. One way to limit the carbon footprint is to shift the transport of battery products from road to rail. Focusing on the design of the transport system for battery systems and cell modules, our design starts with the special load carrier (SLC), as both goods are class nine dangerous goods according to the Agreement concerning the International Carriage of Danger-ous Goods by Road (ADR). Based on the results of a functional analysis, the racks should be designed to fulfil the specific load carrier functions for cell modules and battery systems. The combination of SLC design and cargo space optimisation results in the best possible use of the volume and mass capacity of freight wagons for intercompany transport by rail. Special trailers are used for internal transport with electric trucks. All loading/unloading, handling, storage and transport processes are fully automated to ensure effi-cient internal material flow. An application example shows that around 14,300 tonnes of CO₂e can be saved per year for the supply chain of cell modules and battery systems by rail. This result points the way to further practices and improvements in the sustainability of automotive or other sup-ply chains.

In a dynamic and uncertain global environment, supply chain reconfiguration has emerged as a solution for managing disruptions and enhancing operational resilience. This paper presents an intelligent reconfiguration framework developed following an in-depth analysis of the existing literature on reconfiguration, the application of knowledge graph and the integration of Artificial Intelligence in supply chain reconfiguration. This comprehensive literature review highlights gaps in current methodologies, particularly in achieving efficient reconfiguration. In response, we propose a novel framework that leverages real-world data, Digital Twins modeling, Dynamic Knowledge Graph and Artificial Intelligence-driven Decision Making to enable intelligent supply chain reconfiguration. This framework facilitates the continuous evaluation and optimization of supply chain configurations, while incorporating key performance indicators such as sustainability. By using the Artificial Intelligence to enhance predictive capabilities and decision-making, the framework ensures that supply chains can seamlessly adapt to disruptions and evolving requirements. This work is driven by the need for sustainable and resilient supply chain that maintain a competitive edge in dynamic environments, a goal that can be achieved through efficient dynamic reconfiguration.

Technology integration remains a critical challenge for organizations striving to maintain competitiveness amid rapid technological advancements and evolving market conditions. Despite its potential for enhancing productivity and strategic growth, organizations often struggle with selecting and aligning appropriate technologies with their core competencies and long-term objectives. This study presents a conceptual framework that facilitates structured decision-making by aligning organizational capabilities with strategic objectives in technology integration. The research employs a Design Science Research (DSR) approach, incorporating a pragmatic paradigm and elements of design thinking to develop a robust framework. Empirical data were collected through semi-structured and unstructured industry interviews, complemented by a systematic literature review sourced from reputable academic databases. The conceptual framework integrates key theoretical perspectives from operations strategy, innovation management, technological intelligence, sustainability, risk, and stakeholder management. Additionally, strategic assessment tools such as gap analysis, impact/effort analysis, and technology screening mechanisms contribute to the framework’s methodological soundness. The framework was empirically validated through case studies in manufacturing companies, demonstrating its effectiveness in guiding technology selection and integration decisions. Key findings indicate that the framework provides a structured, visual, and analytical tool to assess technology projects based on strategic alignment and organizational capabilities, enabling informed decision-making. Industry feedback highlighted its relevance in cross-functional collaboration and its potential for adaptation to diverse industrial contexts. Future research will focus on refining the framework’s adaptability across different industries and enhancing its application in project portfolio management. The study underscores the importance of a strategic, capability-driven approach to technology integration, ensuring sustainable competitive advantage and long-term business value.

Available locations for new factories and the suitability of these locations chang-es, especially due to climate change. Long-term consequences and high costs en-tailed by factory location selection become more difficult to predict. The decision process for factory location therefore becomes more complex, uncertain and time-consuming. We aim to develop a data-driven digital decision support application for factory location selection. This paper presents the process to be implemented in that system. We research and refine established processes to enable data-driven, objective and sustainable factory location selection. Finally, we give an outlook on the decision support system to be developed.

Small and medium-sized enterprises (SMEs) are vital to the economy across various business domains, including manufacturing. However, as manufacturing principles and technology evolve, SMEs often struggle to benefit from new knowledge and technologies fully. Their lacking capacity to absorb and use this expertise in physical and data processes hinders manufacturing innovation. One way to obtain this capac-ity is through professionals' engagement in learning-factory activities. Learning facto-ry approaches have been proposed to facilitate the knowledge and competence de-velopment often needed for manufacturing innovation. To advance the understand-ing of the effect of knowledge and competencies gained through Learning Factory approaches on Manufacturing Innovation, this study explores the question: “What role do diverse learning factory approaches play in facilitating SMEs’ transition to new manufacturing technologies and operational strategies?

Smart product-service systems (sPSS) revolutionize mass customization by adapting solutions to changing customer requirements, yet demand substantial engineering to ensure valid future configurations. A "maximum configuration" approach installs all potential features regardless of use. While enhancing scala-bility, this raises concerns about sustainability. Traditionally, sPSS address de-fined customer problems within optimized supply networks, but especially busi-ness ecosystems require broader perspectives regarding complementarity. This paper introduces compound features, derived from systems engineering, as emergent properties of interactions between system components. By treating compound features as resources and capabilities and employing a resource-based configuration strategy, the paper highlights how this approach helps to find capability gaps and align requirements across entities. Furthermore, it discusses compound features within different hierarchical levels of business ecosystems.

The increasing complexity of product configurations demands intelligent systems that effectively integrate customer requirements, dependencies, and uncertainties. This paper introduces a unified framework combining Knowledge Graphs (KGs) and Bayesian Networks (BNs) to improve the efficiency and adaptability of product configuration processes. KGs provide a semantic foundation for product information, ensuring interoperability and clear modeling of relationships. BNs enhance this through probabilistic reasoning, allowing the system to manage uncertainties and dynamically generate optimal configurations . The integration of deterministic, rule-based reasoning from ontologies with the probabilistic nature of BNs automates suggestions, predicts user preferences, and simplifies complexity. This framework streamlines user interactions through intelligent form pre-filling and contextually relevant suggestions, even under uncertainty. By employing an ontology-based representation of BNs, the components fit seamlessly into the KG, creating a cohesive and unified framework that balances scalability and user-centric design to address modern configuration challenges.

In recent years, product portfolio management (PPM) has faced increasing pres-sure from various stakeholders due to the sustainability transformation. This re-sulted in a higher complexity of decision-making and necessitates the systematic integration of sustainability into PPM to achieve corporate sustainability targets. However, incorporating this new dimension introduces target conflicts as well as a significant data demand. This paper presents a concept for the sustainability-oriented variant management of technical products. The concept enables portfolio managers to define, streamline and operationalize the sustainability controlling in accordance with the company’s target system. Furthermore, it fosters the scalabil-ity of sustainability assessments of products with machine learning approaches. This addresses the industry’s need for faster and more practical assessment alter-natives with sufficient preciseness to provide transparency for decision-making in multi-variant product portfolios. Finally, the concept supports users in linking the insights with measures of variant management, thus promotes the sustainable de-velopment of the product portfolio.

In observing the evolution of product configurators since 2007 in our Configurator Database Research Project we have been conducting analyses on the development of configurators based on various criteria as well as exploring the appearance of their user interfaces. Being a crucial tool in the concept of mass customization the criteria for successful product configurators have changed and adapted to the evolving needs of customers and the technological possibilities. The rise of Artificial Intelligence (AI) is and will massively influence the way how software interaction is taking place. In our paper we examine if AI will augment and enhance a product configurator experience and what features and effects might change in how configurator interaction will take place in the future

Knowledge-based CAD systems provide innovative solutions for automating de-sign processes and enabling efficient, user-centric product customization. This paper presents a CAD configuration system for designing subway seat layouts, where users can input parameters such as train length, number of seats, and com-fort settings. The system outputs a complete seat layout along with the maximum and standard seating capacities. This approach facilitates rapid visualization and evaluation of seat configurations, offering a valuable tool for generating and re-fining design concepts. The study outlines the development and implementation of this knowledge-based approach, focusing on the integration of parameter con-trol into the CAD environment using Microsoft Excel and Autodesk iLogic. By externalizing parameter control, the system simplifies the process of visualizing and testing layout ideas. The results demonstrate the system's ability to handle complex interdependencies between design variables while providing intuitive customization options while keeping efforts for adaptation and maintenance of the system low. The study outlines the development and implementation of this knowledge-based approach, focusing on the integration of parameter control into the CAD environment using Microsoft Excel and Autodesk iLogic. This integration enables swift iterations and adjustments, significantly reducing the time required for design and construction. By externalizing parameter control, the system simplifies the process of visualizing and testing layout ideas, making it accessible to designers and stakeholders. The results demonstrate the system's ability to handle complex interdependencies between design variables while providing intuitive customization options. By comparing this system to traditional design workflows, the study highlights its effectiveness in streamlining the design process, improving adaptability, and enhancing decision-making.

The aim of this study is to examine the cooperative activity of two people who work together to transport a heavy box in the same way. To do this, an experiment was set up with different positions at and around a table using a heuristic approach. The aim was to find out what significant differences in posture and muscular strain exist between the two subjects. The movement and muscle activity measurements focus on the level of differences in the stress on the right hand-arm system. Finally, the causes of these differences are discussed and a hypothesis is formulated for further research into the relationship between muscle activity and human support behavior.

Flexibility and individuality in product portfolios are increasingly important in manufacturing, necessitating highly adaptable production systems. Industrial robots, despite their potential, are often underutilized due to limited autonomy and programming challenges. This paper presents a modular software toolbox designed for cognitive robotics while addressing the research question of how a modular software toolbox can enhance the cognitive capabilities and autonomy of industrial robots to improve adaptability and efficiency in automated manufactur-ing processes, particularly in the context of small to medium batch inspections, alongside the increase in efficiency during commissioning and application. The toolbox features reconfigurable algorithms for object recognition, handling, and quality inspections using machine learning methods such as CNNs and SVMs. Optimized for standard industrial computers, the system ensures efficient resource use, achieving object recognition at 20 frames per second. The setup includes a six-axis robot with a CNC controller and highprecision sensors, offering significant cost and time savings for small- to medium-batch inspections. Validation demonstrates the toolbox's potential to enhance flexibility and intelligence in manufacturing systems.

Advanced digital technologies are developed and implemented to address challenges in complex manufacturing systems. Two well-known technologies in this domain are virtual commissioning and discrete-event modeling and simulation. These simulation tools are supported by well-substantiated theoretical frameworks and proven in industrial applications. However, the two approaches target different aspects of production systems and their combination is not extensively researched, thereby motivating this study. This paper reviews existing literature on the combination of these simulation approaches, aiming to identify current opportunities, challenges, and research gaps, while proposing innovative directions for future exploration informed by the findings. The results suggest future research should explore combining VC and DES to enable test-driven development, reduce modeling effort through joint model design, and extend simulation model usability across the production system life cycle.

The application of augmented reality (AR) has the potential to transform industrial processes by integrating digital information seamlessly into the physical environment. AR systems are widely employed in various applications including assembly, quality control, and inspection, thereby offering significant opportunities to enhance operational efficiency. AI-driven features facilitate real-time image validation, representing a novel advancement in the field of AR validation. Nevertheless, these features also raise crucial concerns regarding their practical implementation and associated challenges. Despite the growing use of AR in industrial contexts, scientific studies addressing practical experiences and challenges remain scarce. Therefore, this paper presents a detailed analysis of the key challenges and benefits associated with the integration of AR systems into industrial workflows. The analysis identifies critical issues, such as user acceptance, system reliability, and the alignment of AR systems with operational requirements, while also highlighting advantages like in-process employee training. Based on these insights, actionable recommendations are provided to overcome barriers and reduce adoption hesitations. These include recommendations for tailoring AR systems to specific operational needs and fostering user engagement through targeted training programs. The findings emphasize the need for iterative development processes that adapt AR systems to dynamic industrial contexts, ensuring long-term usability and scalability. Moreover, the paper underscores the importance of user-centered design in enhancing operational acceptance and achieving tangible benefits. These measures pave the way for a more effective and sustainable integration of AR into real-world manufacturing applications, driving the broader success of digital transformation initiatives.

A Digital Human Model (DHM) enables computer-aided analysis of production processes and supports the ergonomic design of workflows. In particular, DHMs are often used to analyze manual work and evaluate ergonomic parameters. Previous studies show that the hand models for ergonomic evaluation are still limited due to inaccurate movement patterns. This study focuses on the simulation of manual assembly work using an anthropometric model. It illustrates with motion measurements how large the angular deviation of the hand postures is. To investigate ergonomic parameters, a cordless screwdriver was used to assemble screws on differently angled mounting plates, which caused different hand-arm postures. The motion capture was done with IMU sensors. Fifteen male subjects aged between 20 and 30 took part in these motion measurements. The results show the effect of the inclined position of the mounting plate on the wrist angles of the respective test subjects. Finally, the deviations between simulation and measurement are discussed and the limitations are shown.

A well-designed layout is essential for efficient production, which significantly contributes to a manufacturing company's economic success. In the era of mass customization, production planners must adapt resources in the layout to accommodate various product variants. However, they face constraints in time and resources, limiting their ability to explore the complete solution space for an optimal configuration. Meta-heuristics like genetic al-gorithms and simulated annealing frequently produce suboptimal solutions and are susceptible to parameter tuning. This paper presents a novel layout optimization approach that converts the layout problem into a mass-spring-damper system, where the equilibrium state represents the ideal layout for resource arrangement. It applies this approach to a brownfield layout problem involving 20 resources, utilizing robots as the handling mechanism. It analyzes the results, assesses the performance of the method, and discusses its limitations. Finally, it outlines potential avenues for future research to address these limitations.

Modular product structures are a proven design method for offering a wide variety of customized products to the market while maintaining a low internal product and process complexity and creating several cost benefits for product manufacturers throughout their value chain. However, the economic benefits of modularity-induced effects have proven challenging to quantify using traditional costing methods. This leaves modular product structures at a disadvantage during the product development phase. Since it comes with a high upfront development cost, the economic benefits gained later are challenging to assess accurately. Effecting the overall product design process. This study aims to provide an overview of the cost-allocating methods currently used to evaluate the economic benefits of modularization. It will examine the existing cost accounting methods for the product life cycle's development phase. It will contribute to a broader understanding of current methods' capability and enlighten potential issues and opportunities in existing practices. Finally, opportunities for improving existing resource allocation methods, which can potentially refine cost allocation systems, are discussed. This would improve the accuracy and detailing of cost when evaluating product portfolios with highly complex, customized, or modularized products and function as decision support for design engineers when evaluating existing products in the design process.

Assembly assistance systems are widely used in industrial assembly to help man-age complexity due to various product variants and to avoid quality issues. The effectiveness of these systems heavily depends on clear and precise assembly in-structions. However, these instructions are often still created manually, which can lead to errors, and the setup process can be very time-consuming. To improve the setup process for informational assistance systems, this article proposes a novel system for the automated generation of assembly instructions using the concept of teaching by demonstration. The foundation of the developed system lies in the application of state trees, which interpret the assembly process captured by cam-eras and convert it into an organized assembly sequence. The system also in-cludes a human-machine interface that enables additional input options for further enrichment of the assembly instructions with context-based information. The as-sembly instructions created can be transferred via an interface to various assem-bly assistance systems for assembly process execution. The developed system was evaluated by a user study, which revealed both its potential for reducing the time needed to set up assembly instructions and needs for further research.

In many small and medium-sized enterprises (SMEs), assembly workplaces often evolve over time without structured planning. This frequently results in inefficient workflows and suboptimal workplace layouts. A primary reason is the complexity of existing optimization methods, which typically require extensive expert knowledge and are therefore challenging to apply in SMEs. This paper presents the development of a user-friendly methodology for the semi-automated determination of optimal assembly workplace configurations. The methodology integrates both assembly time- and ergonomics-based optimization criteria, allowing both to be considered simultaneously. By incorporating both aspects within the optimization process, the methodology identifies configurations that maximize efficiency while minimizing physical strain on employees. To support data collection in non-digitized workplaces, an augmented reality (AR) application was developed for the Microsoft HoloLens 2. This application assists workers in capturing critical workplace parameters, including component distances, movement sequences, and process flows. The methodology analyses the workplace using a structured combination of primary-secondary analysis, Methods-Time Measure-ment Universal Analyzer System (MTM-UAS), and a variant of the Rapid Upper Limb Assessment - NEPRA. Afterwards the workplace is optimized by systematically changing the arrangement of all parts containers. The methodology also guides the user through the optimization by providing visual and textual instructions based on the results of the analyses. The proposed methodology provides companies, especially SMEs, with a practical tool for systematic workplace optimization without requiring extensive prior expertise.

Uncertainties and disruptions across value chains, from supply chains to shopfloor resources, significantly impact production systems. Addressing these challenges requires manufacturing systems that are both resilient and reconfigurable to maintain robust and flexible operations. This work presents a conceptual framework to improve the resilience and reconfiguration capabilities of manufacturing systems at the machine level. The resilience quanti-fication and reconfiguration support are enabled by dataspaces and the Asset Administration Shell (AAS). The goal is to facilitate shopfloor workers in measuring resilience and implementing reconfiguration strategies effectively in response to disruptions. This framework is designed with the focus on progressive forming press machines potentially supporting digital manufacturing. The findings reveal the possibility for production systems to be adaptable and robust. Future work will focus on implementation and evaluation of the proposed framework and scaling it to other manufacturing processes.

Uncertainties and disruptions across value chains, spanning from supply chain to shopfloor at resource level, highly affect production. To ensure robust and flexible production, resilient and reconfigurable manufacturing sys-tems are required. The work presents a service for the resource level, in this case multi-stage forming tools, which recognizes fault situations and suggests actions for reconfiguration to the operator in order to react appropriately to the situation. The methods behind the service are fuzzy logic and case-based reasoning. The utilisation of machine and sensor data, in conjunction with expert knowledge, which will be employed in the construction of member-ship functions, fuzzy rules and the CBR, facilitates the early detection, identification and removal of faults in the production process. The performance of the service is quantified using a benchmark. In addition, an outlook is given on the effects of the applied method on robustness and flexibility of the manufacturing system.

Modern make-to-order (MTO) job shops are vulnerable to unanticipated disruptions, ranging from machine failures to supply chain fluctuations, threatening their ability to meet delivery commitments. This study introduces a framework designed to enhance operational resilience by integrating the Relative Lateness Proportion (RLP) to quantify delay propagation, along with the Active Period Percentage (APP) for monitoring productive capacity. These metrics are combined with an advanced Attention-based Long Short-Term Memory (A-LSTM) prediction model. Validated through extensive testing in a real-world precision manufacturing facility, the proposed solution demonstrates significant advantages over conventional methods, including Autoregressive Integrated Moving Aver-age (ARIMA), vanilla Recurrent Neural Network (RNN), and basic LSTM architecture. The framework enables the early identification of vulnerable work-stations, accurate prediction of disruption patterns, and data-driven interventions to maintain production schedules. Its performance is attributed to integrating operational metrics with machine learning, where the attention mechanism effectively prioritizes critical production data. This research provides manufacturers with a practical, scalable approach to improving delivery reliability in complex, high-variability production environments.

Companies that deliver customized solutions through projects are interested both in the efficient resource use in their manufacturing process and complementing a core product with services to fulfill customers’ specific needs, that is, efficiency value and service value. Firms face tensions, trade-offs, and synergies in aligning efficiency value and service value. This study focuses on the development and delivery of customized solutions in business-to-business (B2B) settings, aiming to uncover project actors' perceptions of efficiency value and service value and the mechanisms used to balance them. A qualitative exploratory study conducted in two software companies and two shipyards reveals that service value and efficiency value are perceived through three aspects: benefits and sacrifices (from both short-term and long-term perspectives), related lifecycle phases and activities, and various enablers. Ten balancing mechanisms harmonize the tensions and trade-offs between these two values. Internal balancing mechanisms include standardizing, modularizing, configuring, reusing, and scaling. External balancing mechanisms comprise co-creating, negotiating, adapting, bargaining, and segmenting. These findings enhance our understanding of value perceptions from a dual-lens perspective and illuminate balancing mechanisms in project business and solution delivery. The examination of software firms and shipyards broadens mass customization research by providing empirical evidence from unconventional contexts.

In multi-variant series production, companies face the challenge of ensuring factors like flexibility and efficiency in planning and ordering despite external and internal turbulences, while simultaneously guaranteeing production as well as procurement stability. To address this challenge, the concept of planned orders has the potential to holistically optimize over the entire planning and ordering process from generating fully specified product configurations, followed by the scheduling of these to valid virtual production programs, until the assignment of incoming customer and stock orders. In doing so, precise material requirements forecasting and simulation-based analyses of future production scenarios are directly integrated in the order-to-delivery (OTD) process, facilitating the early identification and avoidance of potential bottlenecks and risks. This paper introduces a data-driven approach that optimizes the distribution of build-to-stock (BTS) orders across the global sales network, enabling the recommendation of demand-oriented product configurations to dealers. Therefore, several data sources such as historical orders, product structure and market as well as dealer information are taken into account to train the developed algorithm. Coupled with the consideration of existing constraints and given uncertainties, the method offers the ability to allocate pre-planned orders across the dealer network, with the objective to align future customer demand optimally. The presented approach is validated by a real-world use case of the Dr. Ing. h.c. F. Porsche AG to demonstrate the potential of significantly improving the allocation of orders to the dealers in terms of fulfilling upcoming customer requirements, while guaranteeing stable production and procurement processes.

Extending the lifespan of clothing by strengthening the user-object bond is becoming a growing subject of research to reduce the environmental impact of the fashion industry. Meanwhile, mass customization, based on the use of flexible manufacturing systems and digital configurators, enables the creation of garments which are tailored to individual preferences while maintaining production efficiency, which seems to be an interesting avenue to explore a long lasting the user-object relationship. This study assesses the economic, and organizational implications of integrating mass customization into current value chains to reinforce emotional durability. It also critically examines the operational challenges, such as the logistical complexity and carbon footprint of customized production. This empirical analysis is based on a survey conducted in March 2025 with ten Quebec-based fashion companies involved in or exploring mass customization. Results indicated a strong perception of sustainable potential, especially regarding product durability and inventory efficiency. Moreover, there is limited consumer demand for personalization despite interest. These findings highlights the need for improved communication and operational streamlining. By focusing on the evolution of business models, this paper aims to inform both researchers and practitioners about the transformative potential of these approaches and proposes concrete pathways to support their adoption in the fashion industry.

Mass customization in the fashion and apparel industry falls between two extremes: on one end, highly personalized designs tailored to individual needs, and on the other, standardized production aimed at efficiency and profitability. Through a qualitative survey, this article explores how hybrid customization production models can bridge this gap by catering to specific consumer demands while maintaining operational flexibility. The findings suggest that these models enable a two-way interaction between brands and consumers, allowing for continuous adjustments to better meet individual preferences. However, they also place pressure on value chains, particularly in areas such as human resource management, material procurement, and production planning. A conceptual framework specific to the fashion industry is introduced to highlight the key challenges to consider, as well as the opportunities from operational, consumer, and sustainability perspectives. The study concludes with recommendations for implementing hybrid models in fashion, such as combining standardized base production with strategic customization options. These approaches have the potential to enhance customer satisfaction, improve profitability, and reduce environmental impact.

Recent disruptions to the global supply chain have increased maintenance organizations’ focus on spare parts supply resilience and securing the continuous supply of critical spare parts for their production plants. Studies have identified several characteristics of spare parts that challenge planning for spare parts availability. Studies also indicate an industry need to reduce the technical expertise required to implement theoretical models for the management and alignment of spare parts supply capabilities and equipment coverage. To address these challenges, this study proposes a systematic, data-driven portfolio assessment approach to conducting risk-based assessments of spare parts supply capabilities and spare parts coverage of the maintainable asset. The approach is tested in a case study on a scope of 15,427 spare parts in a major offshore oil and gas company. The case study demonstrates enhanced decision-making due to the improved, risk-based overview. The approach also requires low modelling competencies and can be modelled in business intelligence software utilizing existing computerized maintenance management system data. Practitioners can use existing data to assess current spare parts coverage and strategically adjust supply chain capabilities to reach the desired portfolio coverage.

Modern manufacturing environments require adaptable, low-cost vision systems, particularly for robotic end-effectors that must handle a diverse range of workpieces. This study evaluates the ESP32-S3 microcontroller integrated with the OV2640 camera as a compact, wireless vision module compatible with ROS2. The system transmits JPEG-compressed images over Wi-Fi using the micro-ROS communication stack. Performance was analyzed across various image resolutions, and the video stream was processed on a remote ROS2 host using OpenCV and YOLOv8n. To identify performance bottlenecks, high-resolution timing measurements were conducted along the data pipeline. Results show that image acquisition and message preparation contribute negligible delay (<0.1 ms), while most latency arises during image encoding, serialization, and wireless transmission, with a median delay of approximately 78 ms. This limits the achievable frame rate between 18.3 and 40 frames per second. The findings demonstrate the ESP32-S3’s suitability for motion detection and basic object recognition in low-cost, real-time robotic applications.

Manufacturing companies face a dynamic environment that demands frequent adaptations of production processes and infrastructure. These so-called Manufacturing Changes (MCs) differ in scope, complexity, and cost, making standardized approaches insufficient for addressing specific company needs. Effective decision making requires tailored solutions that consider both company-individual and change-specific requirements. This paper presents a software tool that systematically supports MCs by analyzing upcoming changes and applying a correlation model to derive necessary actions. The tool integrates change characterization with decision-support functionalities to enhance the effectiveness and the efficiency of MC handling. Developed as a web-based application using HTML, CSS, JavaScript, and Python, the tool provides actionable insights to optimize decision making. Initial tests demonstrated its potential to support adaptive manufacturing environments, enabling companies to respond proactively to production changes. This solution represents a significant step forward in managing MCs by combining a scientific methodology with a practical application to enhance operational flexibility.

To support high-speed mechanical search and self-aligning insertion of parts with tight tolerances and sensitive components, compliance directly proximal to the gripped object has been proposed in previous work, provided by structured compliance in additively manufactured grippers. The grippers are designed parametrically via the slicer software, allowing a quick iteration over the design parameters. Due to the design process, each combination has to be remodeled manually in CAD in order to be analyzed in simulations. This work focuses on the current modeling problem, by proposing a python script to directly abstract a solid from g-code. The object reconstruction runtime and quality are compared to other state-of-the-art approaches achieving reconstruction up to 97.9 \% times faster for objects with a higher complexity by also achieving a high reconstruction quality. The reconstruction time can be accelerated further in future work, by implementing a script-based modeling.

The use of natural markers can improve the consistent and forgery-proof traceability of assets in sustainable supply chains. When using fingerprint technology, the inherent surface structure of an asset is captured by an imaging system and then compressed into a digital asset fingerprint which is stored in a database. To identify an asset, a newly generated fingerprint is compared against those in the database. This approach is particularly relevant for enhancing consistent asset traceability to meet regulatory requirements and support initiatives such as the digital product passport. To consistently locate and align an asset’s fingerprint region of interest, this work investigates the use of bounding symbols. Five bounding symbol shapes are empirically evaluated for their recognition performance. Additionally, four open source algorithms (ORB, BRISK, SIFT and pHash DCT) to create asset fingerprints are compared and evaluated in terms of identification confidence, processing speed, and memory requirements to assess their feasibility for tracing assets throughout life cycles. The experimental setup examines aluminum raw castings in the form of medallions. Results reveal differences in the identification confidence and resource usage among the tested algorithms. Notably, pHash DCT is more than three orders of magnitude faster than the algorithms used for feature matching, requires the least storage space, and still provides sufficient identification confidence.

In robotic surface finishing, ensuring feasible contact force during defect removal is critical for maintaining grinding quality and tool safety. This paper presents a force controlled pose adaptation method for robotic grinding. In this respect, success of grinding process, defect removal speed and surface quality are investigated to evaluate the proposed method. The control system employs real-time force feedback to adjust the TCP trajectory, moving backward when excessive force is detected to maintain contact force within a feasible range. Experiments of grinding tasks on zinc-coated scaffolding demonstrate that the proposed method enables successful defect removal while maintaining force levels.

The digital transformation of industry, also known as Industry 4.0, relies on various technologies within manufacturing, data processing, and sen-sors. One of these technologies is machine vision, which allows real-time quali-ty inspection. However, small and medium-sized enterprises (SMEs) struggle to adopt this technology. Hence, an overview of this technology's potential barri-ers and opportunities for adoption is needed to create awareness. A deeper un-derstanding of these barriers will help companies overcome these challenges. This study identifies adoption barriers and opportunities through stakeholder interviews based on empirical data from three enterprise classes: machine vi-sion manufacturers, Danish SMEs manufacturing and engineering services, and producing. The interviews are analysed through the Gioia methodology and are divided into sensing, seizing, and transformation barriers among the enterprises. Next, some of these barriers are further investigated by analysing automation engineering students' experiences and observations during their internships. Additionally, we examine how these barriers are represented among the enter-prise classes. Identifying these barriers can assist practitioners and researchers in future work and progress within the field.

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 screw runs 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\% improvement over benchmark), while PAA offers the best balance between performance and computational efficiency. For minimal resource usage, catch22 maintains performance within ±6\% of benchmark using only 24 features. These findings provide concrete guidelines for selecting feature reduction techniques based on application requirements.

Surface defects in galvanized scaffolding such as excess zinc on the metal surface can significantly affect assembly precision and durability. Traditional inspections based on manual visual assessment are limited in efficiency and cannot be easily automated. In this work, a 3D scanning-based method is proposed for automated defect detection in galvanized scaffolding. A 3D scanner is employed to acquire point cloud data of scaffolding components. The scanned point cloud is then aligned with an uncoated 3D Computer-Aided Design (CAD) reference model using the Iterative Closest Point (ICP) algorithm. Defects are identified by analyzing spatial deviations between the scanned data and the reference model. Deviations are considered as defects when exceeding a predefined threshold. To eliminate false positives defects and deviations at irrelevant areas, two region-specific filters are applied. Experimental validation demonstrates the improved automation of the proposed method compared to manual visual inspection.

Robots today still struggle with adaptation and generalization to changes in the task, a major barrier to deploying robots in semi- and unstructured tasks. If robots can detect when novel situations are encountered, they can take a fail-safe or fallback action to recover or at least avoid damage. Anomaly detection (AD) identifies data patterns that deviate from expected behavior. We apply a variational autoencoder approach to time series in robotics for an industrial cabling task. Inputs are force measurements and the robot's end-effector positions, from both nominal processes and various failure scenarios. In validation, the AD model achieved an AUROC of 0.93 in detecting a process-related failure. In the overall evaluation, two of three types of failures were reliably detected, while the third, which had smaller magnitude deviations in the force profile, proved challenging to identify robustly.