Failure mode and effects analysis (FMEA) is a critical but labor-intensive process in product development that aims to identify and mitigate potential failure modes to ensure product quality and reliability. In this paper, a novel framework to improve the FMEA process by integrating generative artificial intelligence (AI), in particular large language models (LLMs), is presented. By using these advanced AI tools, we aim to streamline collaborative work in FMEA, reduce manual effort and improve the accuracy of risk assessments. The proposed framework includes LLMs to support data collection, pre-processing, risk identification, and decision-making in FMEA. This integration enables a more efficient and reliable analysis process and leverages the strengths of human expertise and AI capabilities. To validate the framework, we conducted a case study where we first used GPT-3.5 as a proof of concept, followed by a comparison of the performance of three well-known LLMs: GPT-4, GPT-4o and Gemini. These comparisons show significant improvements in terms of speed, accuracy, and reliability of FMEA results compared to traditional methods. Our results emphasize the transformative potential of LLMs in FMEA processes and contribute to more robust design and quality assurance practices. The paper concludes with recommendations for future research focusing on data security and the development of domain-specific LLM training protocols.

Generative Artificial Intelligence (Generative AI) is a collection of AI technologies that can generate new information such as texts and images. With its strong capabilities, Generative AI has been actively studied in creative design processes. However, limited studies have explored the roles of humans and Generative AI in conceptual design processes, which leaves a gap for human–AI collaboration investigation. To address this gap, this study attempts to uncover the contributions of different Generative AI technologies in assisting humans in the conceptual design process. Novice designers were recruited to complete two design tasks in the condition of with or without the assistance of Generative AI. The results revealed that Generative AI primarily assists humans in the problem definition and idea generation stages, while the idea selection and evaluation stage remains predominantly human-led. Additionally, with the assistance of Generative AI, the idea selection and evaluation stages were further enhanced. Based on the findings, we discussed the role of Generative AI in human–AI collaboration and the implications for enhancing future conceptual design support with Generative AI’s assistance.

The literature suggests that product and product-service system (PSS) design problems are characteristically different. However, there is limited empirical evidence to suggest that the design cognition specific to the respective design activities is different. This article reports the findings of a comparative study of protocols of conceptual product and PSS designing carried out in a laboratory environment by 28 pairs of experienced product designers from the manufacturing industry. First, differences between product and PSS design problems were theoretically characterized in terms of their respective sources of complexity. Based on these differences, hypotheses concerning differences in the cognitive processes of conceptual product and PSS designing were developed and empirically tested. Results indicate that PSS designing by experienced product designers is more problem-focused while product designing is more solution-focused. PSS designing was found to focus more on the design issue function and the design process formulation. Further, PSS designing was more likely to apply a depth-first search strategy, while product designing was more apt to apply a breadth-first search strategy. Results point towards the need to support the analysis of derived behavior of structure and the application of a breadth-first strategy during PSS designing by product designers.

Resource management skills are critical to success during new product development processes. Design processes are ambiguous and complex, and designers often face a scarcity of resources that limits their ability to move the new product development process forward, such as limited financial capital, time or human resources. A team’s ability to use resources effectively may determine their likelihood of success during new product development processes. Technology-based startup teams represent an authentic, unique subset of new product development teams that are trying to bring innovative technologies to market. While prior work has identified salient traits of team members that affect a team’s trajectory, little work has investigated how these traits may interact with each other and how they affect an individual’s ability to manage resources. Using a mixed-methods approach, we leveraged data from 241 startup team members to study the relationship between individual traits, team characteristics and resource management skills. A k-means cluster analysis unveiled two distinct archetypes of startup team members, differentiated by (1) self-efficacy, (2) bricolage, (3) risk propensity and (4) perceptions of psychological safety. Team members with higher levels of these traits exhibited greater resource management skills.

The ideating phase of product design is critical, as decisions made here influence the rest of the product’s lifecycle. Usually, early preliminary designs in engineering are created with pen and paper, which are incompatible with the subsequent digital design process. In an effort to find a modeling tool for early designs that provides the creative flexibility of freehand sketching but also the further processability of digital models, this research investigates natural modeling in virtual reality (VR). To do so, a VR modeling method allowing the intuitive creation of preliminary designs as simplified computer-aided design (CAD) models is presented. The main contribution is the evaluation of this natural VR modeling method against freehand sketching in an extensive user study.

This research proposes a novel conceptual framework that combines the concepts of Human-Computer Interaction (HCI) and Ambient Intelligence (AmI). The proposed framework aims to shed light on the importance of considering the needs and the social interactions of various building occupants in different types of buildings and designing HBI strategies accordingly. Specifically, we take educational buildings as a case that is less explored in the HBI research and apply the proposed framework, investigating how HBI strategies and interactions should be designed to address the needs of students, as primary occupants. Focus groups and semi-structured interviews were conducted among students in a flagship smart engineering building at Virginia Tech. Qualitative coding and concept mapping were used to analyze the qualitative data and determine the impact of occupant-specific needs on the learning experience of students. “Finding study space” was found to have the highest direct impact on the learning experience of students, and “Indoor Environment Quality (IEQ)” was found to have the highest indirect impact. The results show a clear need to integrate occupant needs in designing HBI strategies in different types of buildings. Finally, we discuss new ideas for designing potential Intelligent User Interfaces (IUI) to address the identified needs.

Analysing hierarchical design processes is difficult due to the technical and organizational dependencies spanning over multiple levels. The V-Model of Systems Engineering considers multiple levels. It is, however, not quantitative. We propose a model for simulating hierarchical product design processes based on the V-Model. It includes, first, a product model which structures physical product properties in a hierarchical dependency graph; second, an organizational model which formalizes the assignment of stakeholder responsibility; third, a process model which describes the top-down and bottom-up flow of design information; fourth, an actor model which simulates the combination of product, organization and process by using computational agents. The quantitative model is applied to a simple design problem with three stakeholders and three separate areas of responsibility. The results show the following phenomena observed in real-world product design: design iterations occur naturally as a consequence of the designers’ individual behaviour; inconsistencies in designs emerge and are resolved. The simple design problem is used to compare point-based and interval-based requirement decomposition quantitatively. It is shown that development time can be reduced significantly by using interval-based requirements if requirements are always broken down immediately.

The design provides innovative solutions to problems in the medical field. Collaboration between design and medicine can be fostered in several ways; however, educational programs linking these two academic fields are limited, and their frameworks and effectiveness are unknown. Hence, we launched an educational project to address medical problems through design. The framework and creative outcomes are based on the results of two consecutive one-year programs. The research subjects were 35 participants from three departments. The majority (22/35, 63%) were master’s and doctoral students in design. Eight participants were doctoral students and researchers who volunteered from the surgery, oral surgery, neurology and nursing departments at the Graduate School of Medicine and Hospital. The impact of the program on creativity was evaluated by the quality of ideas and the participants’ assessments. In total, 424 problems were identified and 387 ideas were created. Nine prototypes with mock-ups and functional models of products, games or service designs were created and positively evaluated for novelty, workability and relevance. Participants benefitted from the collaboration and gained new perspectives. Career expectations increased after the class, whereas motivation and skills remained high. A framework for a continuing educational program was suggested.

Text-to-image models are enabling efficient design space exploration, rapidly generating images from text prompts. However, many generative AI tools are imperfect for product design applications as they are not built for the goals and requirements of product design. The unclear link between text input and image output further complicates their application. This work empirically investigates design space exploration strategies that can successfully yield product images that are feasible, novel and aesthetic – three common goals in product design. Specifically, users’ actions within the global and local editing modes, including their time spent, prompt length, mono versus multi-criteria prompts, and goal orientation of prompts, are analyzed. Key findings reveal the pivotal role of mono versus multi-criteria and goal orientation of prompts in achieving specific design goals over time and prompt length. The study recommends prioritizing the use of multi-criteria prompts for feasibility and novelty during global editing while favoring mono-criteria prompts for aesthetics during local editing. Overall, this article underscores the nuanced relationship between the AI-driven text-to-image models and their effectiveness in product design, urging designers to carefully structure prompts during different editing modes to better meet the unique demands of product design.