In the era of digitization and the growing flood of information, the automatic, role-specific identification of information is crucial. This research paper aims to investigate whether the adaptation of LLM is suitable for classifying information obtained from standards for corresponding role profiles. This research reveals that with systematic fine-tuning, prediction accuracy can be increased by almost 100%. The validation was carried out using a two-digit number of standards for three predefined roles and demonstrates the significant potential of LM for labelling content with regard to roles.

The manual execution of failure mode and effects analysis (FMEA) is time-consuming and error-prone. This article presents an approach in which large language models (LLMs) are integrated into FMEA. LLMs improve and accelerate FMEA with human in the loop. The discussion looks at software tools for FMEA and emphasizes that the tools must be tailored to the needs of the company. Our framework combines data collection, pre-processing and reliability assessment to automate FMEA. A case study validates this framework and demonstrates its efficiency and accuracy compared to manual FMEA.

This study integrates surrogate models into combined design optimization of a binder cover, considering production cost and performance constraints. Results reveal that models trained on substantial datasets achieve designs close to the global optimum. Incorporating model variance into constraints prediction in surrogate-based optimization improves robustness and accuracy, especially with noisy functions. This modification enhances the likelihood of obtaining feasible designs, reducing computational demands and showcasing the potential of smaller datasets in predicting local optima.

Digital Twins are perceived differently between and within industry and academia regarding applications and potentials. For this reason, a round table was formed based on the Digital Twin Workshop of the Design Conference 2022. One of the results of this round table is this contribution, which deals with a survey within the industry. The survey captured the understanding of the different roles in the creation and use of Digital Twins, the requirements and hurdles as well as the perception of methodological support. In addition, factors that influence the perception were identified.

This paper presents a novel method for automatic contradiction detection in requirements engineering using a hybrid approach combining formal logic with Large Language Models (LLMs), specifically GPT-3. Our three-phase process detects contradictions by identifying conditionals and pseudo-grammatical elements, and employing LLMs for nuanced contradiction detection. Tested extensively, including on a real-world electric bus project, our method achieved 99% accuracy and 60% recall. This approach significantly reduces manual effort, enhances quality, and is scalable for future advancements.

Natural Language Processing (NLP) has been extensively applied in design, particularly for analyzing technical documents like patents and scientific papers to identify entities such as functions, technical feature, and problems. However, there has been less focus on understanding semantic relations within literature, and a comprehensive definition of what constitutes a relation is still lacking. In this paper, we define relation in the context of design and the fundamental concepts linked to it. Subsequently, we introduce a framework for employing NLP to extract relations relevant to design.

Hands are the sensors and actuators for many design tasks. While several tools exist to capture human interaction and pose, many are expensive and require intrusive measurement devices to be placed on participants and often takes them out of the natural working environment. This paper reports a novel workflow that combines computer vision, several Machine Learning algorithms, and geometric transformations to provide a low-cost non-intrusive means of spatially tracking hands. A ±3mm position accuracy was attained across a series of 3-dimensional follow the path studies.

The Digital Thread is a system that connects different phases of the product lifecycle and the related data across one or more companies in the supply chain. This work aims to develop a graph data model of the Digital Thread, in the context of the vision of polyglot persistence, that interconnects the different phases of the lifecycle and their corresponding data models, processes, and IT systems. This work proposes a Digital Thread Graph that integrates a Digital Model and a derived Digital Twin, using object and relation attributes for view creation and filtering while minimizing redundancy.

Additive manufacturing (AM) became a key technology in the development of innovative products. Advancements have been made to improve economic feasibility. However, ecological sustainability is still an open issue of AM. To improve sustainability, it is crucial to track, visualize, and evaluate emissions along the lifecycle. This paper presents a novel Digital Twin based approach enabling prediction of the product carbon footprint (PCF) and prescriptive measures to improve sustainability. By improving part and process design, a significant PCF reduction was achieved.

Cyber-Physical-Systems provide extensive data gathering opportunities along the lifecycle, enabling data-driven design to improve the design process. However, its implementation faces challenges, particularly in the initial data capturing stage. To identify those, a comprehensive approach combining a systematic literature review and an industry survey was applied. Four groups of interrelated challenges were identified as most relevant to practitioners: data selection, data availability in systems, knowledge about data science processes and tools, and guiding users in targeted data capturing.

The utilization of data in design is a crucial aspect of shaping the product and service development. Despite the lack of extensive research on this subject, this study aims to bridge the gap by introducing the ‘D³IKIT’, a data-driven design process and toolkit. Through workshops, this process and toolkit offer a practical method for creating innovative product and service concepts using data and machine learning. Developed and tested with the participation of 42 individuals, the ‘D³IKIT’ provides valuable insights for both practitioners and academics.

Data-driven design is believed to be empowered by machine learning (ML) with advanced pattern classification and prediction. However, research on how ML can be used to support automotive human-machine interface (HMI) design is lacking. We presented a case study of truck HMI design to understand the current data use and expectations of ML in the design process. Findings show decentralized data practices, the role of expertise in decision-making, and the envisioned reactive use of ML, where we underscore the implications for advancing human-ML collaboration in designing future truck HMI systems.

In the realm of process engineering, the pursuit of sustainability is paramount. Traditional approaches can be time-consuming and often struggle to address modern environmental challenges effectively. This article explores the integration of generative AI, as a powerful tool to generate solution ideas and solve problems in process engineering using a Solution-Driven Approach (SDA). SDA applies nature-inspired principles to tackle intricate engineering challenges. In this study, generative AI is trained to understand and use the SDA patterns to suggest solutions to complex engineering challenges.

The optimization of mechanical behavior in safety systems during crash scenarios consistently poses challenges in vehicle development. Hence, a reinforcement learning-based approach for optimizing restraint systems in frontal impacts is proposed. The trained agent, which adjusts five parameters simultaneously, is capable of minimizing loads on a seen and unseen anthropomorphic test device on the co-driver position and is thus able of transferring knowledge. A hundred times higher rate of convergence to reach a similar optimum compared to a global optimization algorithm has been achieved.

Despite its transformative impact, a systematic approach to Smart PSS development remains elusive. Addressing this, the study introduces a dynamic conceptual model named DHSmart and its accompanying canvas, adaptable to various contexts and technological advancements. Notably, it offers a structured approach to designing ‘Smart’ in Smart PSS, capturing the interplay between data, humans, and smart systems while directing digitalisation that achieves competitive advantage. It also serves as a unifying framework, enabling meaningful interdisciplinary contributions in theory and practice.

Digital Twins (DTs) are intended to be utilized for a wide range of applications, promising benefits like visualization, monitoring, simulation and control of a physical system. Not only the development of a DT for a production facility is a time-consuming task, but also to keep the virtual counterpart up to date in the use phase. In this work, the implementation of an industrial-scale DT of an automotive supplier production site based on a Discrete-Event Simulation (DES) model with self-optimization capabilities for easier maintainability and increased simulation accuracy is presented.

With the swift entry of artificial intelligence (AI) into everyday life, human-product interactions are becoming increasingly complex. We suggest an ecosystem-minded, humanity-centered design approach to better understand this complexity. Simultaneously with the development of interaction types, discussions and developments on theories of mental models are crucial to understanding and improving the nature of these interactions. In this paper, we address the gap in mental model theories and extend Norman's conceptual model at three dialogue levels: dialogue in language, mind, and use.

This paper introduces a new concept for the automation of the assembly planning process, to enable Design for Assembly (DfA). The approach involves the application of reinforcement learning (RL) to assembly sequence planning (ASP) based on a 3D-CAD model. The ASP algorithm determines assembly sequences through assembly by disassembly. The assembly sequence is then used for the generation of subassemblies by considering the product contact information. The approach aims to support the creation of the manufacturing bill of materials (MBOM) by automating the assembly planning process.

Despite technological and medical advances, amputations continue to increase. Amputees face significant challenges when acquiring and using prosthetic devices, challenges which are made worse as their emotional needs, aspirations, mobility, prosthesis requirements and problems change over time. These challenges require custom solutions for each individual amputee, a fact that current amputee centered prosthesis services tend to ignore. The work reported in this paper contributes an AI based Prosthesis Development Service Framework to cater for the current and evolving needs of amputees.

This paper investigates the use of Large Language Models (LLMs) in engineering complex systems, demonstrating how they can support designers on detail design phases. Two aerospace cases, a system architecture definition and a CAD model generation activities are studied. The research reveals LLMs' challenges and opportunities to support designers, and future research areas to further improve their application in engineering tasks. It emphasizes the new paradigm of LLMs support compared to traditional Machine Learning techniques, as they can successfully perform tasks with just a few examples.