Large language models (LLMs) are increasingly used to address real-world design problems, especially during the design ideation phase. Although LLMs hold substantial promise for concept generation, the understanding of how they can effectively assist designers in enhancing the diversity of design concepts is still limited. In this study, we set up different strategies for prompting multiple professional personas to the LLM for design concept generation, including (1) multiple prompts for concept generation in parallel, each with a professional persona, (2) a single prompt for concept generation with multiple professional personas, and (3) a sequence of prompts for concept generation and update, each with a professional persona. We formulate and test several hypotheses on the effectiveness of different strategies. All hypotheses are tested by constructing professional knowledge bases, selecting design problems and personas, and designing the prompts. The results suggest that LLMs can facilitate the design ideation process and provide more diverse design concepts when they are given multiple prompts in parallel, each with a professional persona, or given a sequence of prompts with multiple professional personas to generate and update design concepts gradually.

Recent advances in AI offer promising opportunities for creative design, particularly through the generation of inspirational images. While prior research has explored the general benefits and limitations of text-to-image tools, there is significant potential in overcoming these constraints by investigating agile, multimodal prompting to facilitate more project-appropriate human-AI interaction. We present the development of a system designed to support both text-based and sketch-based image generation, serving as a research artefact for studying creativity support through multimodal Generative AI. The system enables dynamic dialogue interaction and visualization of the respective contributions. This paper focuses on the development of this AI system as a research artefact to enable future research through design, exploring how multimodal prompting can influence the design process.

The complexity and dynamism of global markets is driving the manufacturing industry to evolve from product-based business models to solution providers by integrating smart services. This shift poses challenges, particularly for the organizational design of technical service organizations that deliver these services. Despite the growing importance of servitization, research on the organizational impact of smart services remains limited. Combining a systematic literature review and case study approach with expert interviews, this study examines how smart services affect organizational design of technical service organizations, in the following dimensions: structure, people, processes, and reward systems. The findings offer initial design approaches for manufacturers transitioning to smart service delivery and advance theoretical and practical insights in this field.

The intersection of design and narrative plays a crucial role in shaping meaningful experiences. While narrative experience has been explored in product design, its role in service design remains underdeveloped. This study introduces a narrative-driven service design approach, integrating narrative to enhance user experiences. Using a Research through Design methodology, ten digital service prototypes were developed, embedding “stories of moments of joy” as a design foundation. Findings suggest that starting with narratives fosters deeper emotional engagement and enhances service interactions. Participant feedback highlights how this approach provides an alternative to traditional problem-solving models, emphasizing narrative-driven innovation in service design. By positioning narrative as a central design element, this study contributes to advancing service design methodologies.

sEMG biofeedback therapy can be used to treat arm paresis after a stroke by using surface mounted EMG electrodes to measure muscle activity in the forearm and provide visual feedback to the patient. Since current sEMG biofeedback systems rely on manual placement of a few large electrodes, they cannot be used to discriminate between individual extrinsic finger muscle activities, which is necessary for training everyday hand movements. In this paper, we present our concept for the development of a device that enables the resolution of individual finger activities. We have developed a method that uses and reduces information from large-scale sEMG scans of a person's forearm to identify suitable locations for the strategic placement of a minimal number of electrodes in a personalised forearm sleeve, which is the key component of an effective biofeedback device for everyday hand movements.

Text-to-Image Generative AI (GenAI) platforms offer designers new opportunities for inspiration-seeking and concept generation, marking a significant shift from traditional visualisation approaches like sketching. This study investigates how designers work with text-to-image GenAI during inspiration-seeking and ideation, aiming to characterise designers’ behaviours through designer-GenAI interaction data. Analysis of 503 prompts by four designers engaging in a GenAI supported design task identifies two distinct behaviours: exploratory, characterised by short, diverse prompts with low similarity; and narrowing, characterised by longer, high-similarity prompts used with detail focused variation functions. The findings highlight the value of GenAI interaction data to reveal patterns in designers’ behaviours, offering insights into how these tools support designers and inform best practices.

This study investigates the integration of Large Language Models with the TRIZ to improve problem solving and innovation in industrial product development. By combining the structured problem-solving framework of TRIZ with LLMs to process large amounts of data and generate ideas, this hybrid approach seeks to overcome the limitations of traditional TRIZ and optimize solution generation. In a case study conducted in an industrial setting, the effectiveness of this integration was investigated by comparing team-generated solutions with those derived using LLMs and TRIZ-enhanced LLMs. The results show that while LLMs accelerate idea generation and provide practical solutions, the additional structure of TRIZ can provide unique insights, however depending on the application context.

To meet the upcoming sustainability challenges, aerospace manufacturers need to develop products that both address complex sustainability factors and ensure profitable realization. Furthermore, the sustainability perspective needs to be lifted from focusing on carbon emissions, and broadened to include a system-level socio-ecological view. Manufacturers are thus challenged to balance sustainability, manufacturability, and performance, but lack the methods and tools to make well-informed decisions. We propose a method for conducting multi-domain trade-off studies in the early design phase. A functional architecture modelling approach is utilized to model performance and manufacturing aspects. Together with a relative sustainability fingerprint conducted on design alternatives, design spaces can be explored with respect to performance, manufacturability, and sustainability.

Automating the structuring of Solution Principles within conceptual design is crucial for efficiently covering Function Structures while reducing time-intensive manual processes. Solution Principles are central in bridging functional requirements and technical implementations, yet traditional methods depend heavily on human expertise. To address this, a novel approach leveraging a search algorithm is proposed to automatically identify an optimal set of Solution Principles for a given Function Structure. The approach formalizes the problem and provides rules for the selection and application of Solution Principles. Key components include a function for applying Solution Principles to functions and a heuristic that minimizes principle selection, guiding the search toward optimal solutions. An evaluation shows the potential of this method to reduce time and effort in early product design.

Concept maps have been used to measure student learning and performance comparisons before and after interventions. However, these concept maps may vary in structure and content. The complexity of the maps are difficult to measure when multiple concepts are embedded within individuals nodes. A systematic coding scheme is evaluated for objectivity and reliability using 22 undergraduate researchers. The findings suggest that the coding scheme is suitable and will allow multiple different researchers to generate similar bi-partite graphs from the notional concept maps generated. Additional work is needed to ensure that the semantic content is not invalidated through the coding scheme.

Engineering design tasks are cognitively complex and there is a growing interest in understanding the neurocognitive processes involved in design. Consequently, researchers are increasingly using bio-physical markers such as eye tracking to study design neurocognition. However, these studies are largely correlational, and little is understood about the construct validity of eye-tracking metrics such as fixation durations and saccade frequency. Moreover, these studies rarely account for non-design factors such as neurodivergence (e.g., ADHD) on eye-tracking metrics during design. We aim to examine this research gap through a causal-comparative study with designers with and without ADHD, performing divergent and convergent design tasks. Our findings call for a deeper investigation into the construct validity of eye-tracking metrics while considering a broad range of external factors.

Functional decomposition (FD) is essential for simplifying complex systems in engineering design but remains a resource-intensive task reliant on expert knowledge. Despite advances in artificial intelligence, the automation of FD remains underexplored. This study introduces the use of GPT-4o, enhanced with a proposed Monte Carlo tree search for functional decomposition (MCTS-FD) algorithm, to automate FD. The approach is evaluated qualitatively by comparing outputs with those of graduate engineering students and quantitatively by assessing metrics such as structural integrity and semantic accuracy. The results show that GPT-4o, enhanced by MCTS-FD, outperforms smaller models in error rates and graph connectivity, highlighting the potential of large language models to automate FD with human-like accuracy.

While performing design tasks, engineers rely heavily on their knowledge. However, the expanding knowledge space makes it impractical to perform the design tasks without external inputs. This study explores how AI can bridge the knowledge space expansion gap in design. The study introduces the AICED framework implemented as a web tool Pro-Explora, leveraging advanced multi-agent LLM technology to accelerate early-stage design tasks. Pro-Explora generates professional problem definitions, PDS documents, and unique solution concept images within five minutes, maintaining creative flow. Its effectiveness was validated in a real-life project, with outputs deemed highly relevant by experienced designers. The study highlights the AICED framework’s industry implications, addressing required knowledge. This pioneering study opens new avenues for specific LLM applications in engineering design.

This paper presents a novel framework for Artificial Creativity (AC) in design, emphasizing the co-development of problem and solution spaces. Grounded in cognitive psychology and design theories, the framework leverages advancements in artificial intelligence (AI), particularly generative AI models, to augment human creativity in design. The study identifies four key design spaces—Solution-Knowledge, Solution-Concept, Problem-Knowledge, and Problem-Concept—and defines operators that automate reasonings within and across these spaces. By enabling simultaneous divergence and convergence of problem and solution spaces, it fosters creativity while balancing novelty and effectiveness. This work bridges AI capabilities with cognitive processes of design creativity, laying a foundation for advancing artificial creativity and human-AI collaboration in design.

This study aims to detect the ability of professors to distinguish design assignments generated by students with and without using AI. Ten students were recruited to undertake a conceptual design task twice, one with and one without the help of AI. 105 higher-education associate, assistant and full professors from industrial and product design programmes were recruited to assess the generated designs using a 7-point Likert Scale with nine indexes. The results indicate that assessors have moderate ability to distinguish between design assignments of students using AI and those where students did not use AI. Three cues to suggest the risk of the design assignment is made with AI instead of students who did not use AI were identified. By considering the three cues, lecturers distinguish design assignments generated by students with or without AI.

As conceptual design sketches are a common tool used to communicate between team members, it is important to understand the relationship between sketch characteristics and engineers’ perceptions. This study evaluates 4 line styles common in many sketch techniques: the Single Line, Feathered Line, Heavily Feathered Line, and Variable Line. 40 mechanical engineers ranked preferences of complex engineering products in these line styles and provided adjectives to describe their choices. Results show that engineers preferred the Single Line, which had the common adjectives of clear and professional. The findings suggest that engineers should generally be learning and using a single, uniform and clean line style.

VR sketching tools have matured to a practical level, enabling use across various 3D design disciplines. Studies into VR sketching in design report beneficial affordances but are based on brief testing of tools in simulated tasks. Consequently, there is a knowledge deficit in understanding how to effectively integrate VR sketching into design projects. We address this gap with a case study on the sustained use of VR sketching in 10 automotive concept design projects over 10 months. In analysing designers’ logbooks, which captured design development, and post-study reflections, we show how the affordances of VR sketching outlined in literature manifest in practice. Specifically, we show how and when designers can exploit the precedence of 3D geometry embodied in VR sketches to advance the design process in terms of several dimensions of design fidelity. We highlight where process advantages are realised through (1) increased spatial fidelity, reducing the time required to iterate 2D sketches, (2) operational fidelity supporting dynamic testing of concept functionality via animation and (3) environmental fidelity supporting contextualising components and storytelling. As such, our findings highlight how and when practitioners can realise the comparative benefits of VR sketching alongside traditional sketching and 3d modelling during the concept design process.

The rise of generative artificial intelligences (AIs) has quickly made them auxiliary tools in numerous fields, especially in the creative one. Many scientific works already discuss the comparison of the creative capacity of AIs with human beings. In the field of Engineering Design, however, numerous design methodologies have been developed that enhance the creativity of the designer in their idea generation phase. Therefore, this work aims to expand previous works by leading a Generative Pre-trained Transformer 4 (GPT-4) based generative AI to use a design methodology to generate creative concepts. The results suggest that these types of tools can be useful for designers in that they can inspire novel ideas, but they still lack the necessary capacity to discern technically feasible ideas.

A structurally transforming multi-mode product can realize a changing set of functions across its modes, replacing multiple related products while offering increased cost, space, and time efficiency. However, there is a lack of connected methods that address the additional design complexities due to the product’s physical transformations and the resulting structural component-sharing between modes. A framework, grounded in standard design practice and built upon existing methods, is proposed to help navigate the two most impacted design stages: 1. Problem Definition and 2. Conceptual Design. The Problem Definition stage in this new framework involves identifying the external factor that determines the product’s modes and defining the functional requirements for the modes and transformation methods. The Conceptual Design stage involves iteratively linking conceptualized forms of each mode to adjacent modes through conceptualized transformation methods. The framework is demonstrated in a case study involving the design of a structurally transforming multi-mode piece of children’s furniture that transforms between a cradle, floor seat and a multipurpose toddler step stool. The proposed framework is a promising step toward systematically, cohesively, and comprehensively addressing design challenges during the development of a wide variety of structurally transforming multi-mode products, therefore facilitating better, more effective product design.

This study explores the alignment between automotive design innovation and consumer acceptance, particularly in the context of two significant trends: electrification and digitalization. We probed the acceptance of replacing rear window with wide-angle camera We surveyed 1,546 potential customers from China and Sweden to assess their openness to such a radical design. Findings suggest a consensus on its futuristic appeal but diverge in adoption willingness. The study offers insights to bridge the design preference gap, positioning customer acceptance as key for car design strategy.