Patents contain valuable design insights, yet manual analysis remains time-consuming and complex. This study explores Large Language Models’ capacity to automate patent analysis for engineering design. GPT-5 and Gemini 2.5 Pro were evaluated across Motivation, Novelty, and Key Invention Features using three patents and expert evaluators assessed outputs through Accuracy & Fidelity, Comprehensiveness, and Analytical Depth. Results indicate LLMs demonstrate proficiency in feature synthesis but exhibit inferential limitations in motivation analysis, underscoring the necessity for human oversight.

Generative AI (GenAI) tools are getting more and more integrated into creative workflows, evolving from assistants to collaborators, and reshaping human-AI interactions in the creative process. To better understand the human side of this co-creation, an interview study was conducted with 19 architecture students participating in a GenAI-supported design futuring course. The study identified 18 roles humans and AI can take during co-creation, along with tool-specific variations and insights into emotional dynamics, creative experiences, perceived agency, and control during the design process.

This paper reviews 89 studies on AI in product platform design, further focusing on 21 multi-domain contributions. The dominant archetype is AI as a Tool × Method × Solution Proposal, with AI mainly used for automation and optimization. Collaborative roles remain rare, especially in requirements and architecture phases. Robust evaluation of AI benefits is largely missing, revealing an automation-centric paradigm and key gaps for co-intelligent, cross-domain platform development.

Engineers need to connect knowledge, based on science and technology, with knowledge about humans and society. To operate in a sociotechnical context a variety of different people with different skills are needed. This paper argues that therein lies an opportunity for all who have the skills and interests to find a fulfilling role in engineering that aligns interests in technical task, their role, their identity, their personal strengths and their values, illustrated by women in engineering and sustainability.

This paper examines how ambient airflow, temperature, and humidity impact the print quality of upcycled biomaterials in Direct Ink Writing, and explores strategies for mitigation. A standardized pecan shell flour ink was used with optimized slicing parameters. Experiments in a controlled climate chamber involved sensor logging and statistical analysis. Airflow improved structural stability, overhang fidelity and bridging, but increased Z-axis shrinkage. Higher temperatures slightly improved bridging, while elevated humidity reduced stability and increased sagging, despite small bridging gains.

This work explores Reinforcement Learning (RL) for the circular design of planar truss linkages using available bars and pins. A bipartite graph representation and elementary action formulation enable agents to assemble mechanisms in a physics-based environment. Results for a force-inverter design problem show 98.5% success for fixed-stock training and 66.0% for shuffled stocks. The method demonstrates RL’s potential for inventory-constrained mechanism synthesis, with future work targeting scalable, indexing-invariant architectures and more flexible connection actions.

Traditional design methods fall short for complex socio-technical systems where social and technical elements co-evolve and emergent behaviors resist decomposition. This paper proposes a seven-stage Design for Complexity Framework integrating systems science and design theory. Stages 3–6 form an iterative co-evolution space where modeling, architecture, and stakeholder co-design mutually shape problem and solution development. A healthcare example illustrates how the framework’s co-evolutionary approach addresses coordination failures that purely technical or purely participatory methods miss.

This paper addresses the lack of empirical knowledge on which demographic groups are most likely to use autonomous buses in the Munich Metropolitan Region. We analyze this question through a large-scale online survey capturing demographics, mobility behavior, and accessibility needs. Results show that younger, multimodal, and well-educated individuals form the core of potential users, while older and car-dependent groups remain hesitant. The findings highlight that successful deployment requires inclusive design, improved accessibility, and targeted communication strategies.

This paper reports iterative industrial prototyping of data collection systems for simulating seafood factories. We identify the data necessary to achieve the level of realism factory designers need for effective design exploration, and propose methods to obtain them. Sixteen physical prototypes showed how prototyping shape dynamic requirements in the design process. Findings indicate that models need 3D shape and texture, which can be obtained from smartphone photogrammetry, and bending stiffness and multidirectional friction from cantilever and inclined plane tests.

This paper presents the development and validation of a new information structure for design methods for an enhanced repository of design methods to support design practice and pedagogy. The structure is based on features derived from challenges faced by design students, practitioners, and educators in understanding, using, and teaching design methods, and it is experimentally validated. Results show 81.2% of participants had no difficulty in understanding and using design methods based on the proposed information structure. Participants reported 8 improvements in the structure.

The present paper proposes a framework for translating lost haptic cues of textiles into digital environments with a view to reducing perceptual uncertainty in the context of online shopping. The model’s integration of touch-related attributes and multimodal representations facilitates reliable customer perception, enhances material communication, and guides designers towards informed decisions. The paper outlines the challenges and opportunities inherent in the domain of multisensory digital textile experiences, while concurrently establishing a foundation for future research in this field.

Mixed Reality (MR) prototyping offers significant design opportunities but introduces complexity in prototype specification. This paper presents a card-based design tool to support designers in this specification process. The tool is based on a comprehensive taxonomy of MR prototype fidelity and foundational research into the interplay between, and value of, different physical and virtual characteristics. A validation study demonstrates that the developed tool supports and guides designer reasoning, resulting in higher quality MR prototypes with stronger rationale for their implementation.

The transition to a circular economy requires products that encourage circular consumer behaviour. Despite the central role of designers in this transition, the design for circular behaviour (DfCB) approach remains under-explored. This paper presents a literature-based conceptual model explaining which factors need to be in place, and how they interrelate, in order for designers to facilitate circular behaviours through product design. By pointing out gaps in the current state, future research directions are suggested to foster the establishment of DfCB.

This study proposes a framework to map biomimetic innovation progress along TRLs and identify recurring development patterns. Pilot results reveal stagnation between TRL 2–4, linked to generic upscaling struggles and biomimetic-specific barriers. Emerging hypotheses suggest early onset of upscaling challenges post-POC and influence of biological model knowledge on progress. Study insights open paths for methodological work to bridge POC obtention and validation difficulties, and for further use of the framework on bigger datasets, to build a baseline for biomimetic innovation development.

This contribution addresses the lack of a structured framework for idea workshops in the Integrated Design Engineering (IDE) and resource-constrained settings. The current workshops in IDE are based on general creativity literature rather than on the processes and experiences inherent in IDE. This contribution derives a sequenced mode and integrates proposals to overcome helps to overcome common pitfalls. The sequencing shows best-practice in IDE and enables untrained users to enhance idea quality and process efficiency. The contribution offers a foundation for creativity technique assignment.

This paper explores the application of Web3.0 technologies to provide de-centralised secure, private, and provenance preserving trust networks for society’s increasingly digital design and manufacture workflows. It provides an overview of the key technologies involved and an example of a minimal trust framework required for issuing jobs between actors and machines in a makerspace. A comparison with centralised AM farm platforms is made and demonstrates how Web3.0 can support emergent trust structures compared to fixed centrally managed structures that actors need to agree to.

Industry is experiencing rising thermal loads, so geometries that improve energy transfer are needed. However, defects arising from overhang in additive manufacturing affect the functionality of triply periodic minimal surface (TPMS) based heat exchangers. This study addresses how TPMS superposition affects heat transferring and overhang critical surfaces. The objective is to quantify the functional and manufacturing trade-offs, and to identify the optimal hybrid cells formed from gyroid, Schwarz and diamond units.

Recent advances in machine learning (ML) offer substantial potential for product development (PD), yet adoption remains limited. A crucial step is identifying suitable ML algorithms for a given PD problem, which requires translating domain-specific formulations into appropriate ML tasks. Prior work indicates that LLMs struggle with this step due to insufficient domain knowledge. Therefore, this study investigates whether a domain-specific GraphRAG approach improves model performance by enriching prompts with structured context from a PD knowledge graph.