Design generation using traditional Computer-Aided Design (CAD) tools remains a labor-intensive and manual task. This paper introduces a framework for automating CAD geometry generation using Large Language Models (LLMs) with function calling and agent workflows. The framework enables both expert and novice designers to use textual prompts to automatically generate CAD code. We evaluate it with five LLMs and four agent workflows. The agent workflow incorporating automated visual feedback outperforms the others, especially with multimodal LLMs like ChatGPT-4o. A case study shows its use in topology optimization and additive manufacturing with minimal human input. Remaining challenges include limitations in spatial reasoning, prompt dependency, and workflow adaptability. Future work should focus on improving design-for-manufacturing capabilities, visual tools, and evaluation benchmarking.

Exploring patterns in large text corpus is essential for effective knowledge discovery in research domains. However, machine-driven methods often introduce noise and rely heavily on parameter thresholds. Human expertise is therefore essential for ensuring reliable outcomes. This study conducts a comparative analysis of a classification task performed by both human and computer algorithms. During the task, human experts are asked to categorize a list of abstracts based on their semantic contents, where computer algorithms perform computations, including network analysis and document embeddings, to group the abstracts. The findings show a significant level of disagreement between human and computer-generated clusters, indicating the need for further investigation into the factors influencing community categorization and incorporating more advanced techniques to improve the results.

This paper examines the integration of Life Cycle Assessment (LCA) in the development method of autonomous product-infrastructure service systems, demonstrating the application on the use case for waste management. Integrating LCA in the earlier Phases of development methodology, sustainability analyses identify key environmental hotspots and improvement strategies. Scenario evaluations revealed the potential for energy-efficient operations with reduced emissions through smart infrastructure integration and optimized system designs. Findings underscore the importance of early-stage sustainability assessments and highlight pathways for achieving eco-design goals in urban robotics. This research work provides substantial insights for scalable, sustainable solutions with autonomous product-infrastructure service systems.

Generative artificial intelligence (GenAI) has the potential to further revolutionize Computer-Aided Design (CAD) by recognizing patterns, making predictions, and generating automated design suggestions. This paper presents a systematic literature review that examines the current state of research on the use of GenAI in CAD-based product development. With a focus on 3D modelling, it provides an overview of current approaches, most used datasets and commonly used AI models. Four application areas where GenAI can enhance CAD were derived: Design generation, Design reconstruction, Design retrieval, and Design modification. In total, 47 papers were selected, analysed and categorised.

The aim of this research is to analyze the potential of Generative Artificial Intelligence (GenAI) to support the design process and overcome creative fixation in teams during the initial problem framing, ideation and concept exploration stage. Fixation is a common problem in design, and can be exacerbated during collaborative work due to diverse issues such as team dynamics or perceived hierarchy. Current research is exploring whether AI can help teams overcome this problem or on the contrary, might actually contribute to it. Through a creative ideation workshop with design students, we investigate how AI influences team dynamics as well as the creative results. We propose a conceptual model to work with AI in a team setting.

Products are often optimized for “most likely” conditions, but unexpected variations can render designs ineffective. Using examples from engineering systems, this paper explores the benefits of leveraging non-linear “payoff functions,” where small changes in conditions lead to disproportionate outcomes. By analyzing the direction and curvature of these functions near observed boundaries, designers could gain an understanding of behavior beyond expected ranges. Non-linear modeling can aid in assessing design margins, especially in long-lived systems. Integrating this approach into design processes can be helpful and effective in considering the “preparedness” of a system in the face of unexpected events of different natures.

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.

This paper examines the impact of complexity on New Product Development (NPD) within the context of an Engineer-to-Order (ETO) organisation. A descriptive literature review identified three categories of complexity: organisational, process and product complexity. The influence on NPD performance due to the dimensions contained in these categories are investigated in terms of the Law of Requisite Variety. A case study of NPD at Héroux-Devtek Inc., a landing gear supplier, evaluates these dimensions in practice. The findings reveal that increased organisational complexity often improves NPD performance, while increased process complexity reduces NPD performance. Product complexity evolves from being ‘complex’ initially to ‘complicated’ or ‘simple’ at delivery. Insights into managing these complexities contribute to understanding their role in achieving project success in the ETO context.

This paper, positioned within two universities’ contexts on design education, explores the critical role of awareness in co-design with individuals who have lived experiences. The study introduces a SkillsLab designed to prepare students for managing awkward moments during co-design sessions. A SkillsLab is an intensive learning activity combining hands-on practice, theoretical insights, and practical exercises to bridge the gap between theoretical knowledge on a topic and the real-life application of this knowledge in a project-based setting. The learning activity aims to enhance students’ confidence and skills in navigating awkward moments in co-design. The findings suggest that such educational interventions can significantly improve students’ preparedness for real-world co-design challenges, fostering a more inclusive and empathetic approach to design.

The Consensual Assessment Technique (CAT) is one of the most effective and commonly used design evaluation methods. However, it fails to capture implicit cognitive processes and has mainly been studied in a homogenous design modality. To bridge this gap, the present study investigates the impact of design ideas represented in different modalities (i.e., text-only, sketch-only, text + sketch) on design evaluations for creativity, novelty, and usefulness, and examine human gaze patterns during the evaluation process. Our findings showed that novice raters exhibit higher interrater reliability and greater convergence in visual attention when rating ideas containing sketches compared to text-only design modality, highlighting the value of visual elements in design evaluations.

The history of weed science as a discipline has been a topic of interest for decades, but it is rare for researchers to consider publications prior to the 19th century or that were not focused on North America. In this article, the development of weed identification manuals in early modern England is documented out of two genres of premodern scientific writing: agricultural treatises and illustrated herbals. These two forms of writing intersected in the late 18th century with the publication of Thomas Martyn’s four-volume Flora rustica, an illustrated guide to plants in British agricultural systems. We argue that the key characteristics of modern North American weed identification guides in English (the use of the term weed to categorize plants, descriptions of plant habitats, and the use of detailed descriptions and/or illustrations of plants for identification purposes) originated in these premodern texts.

Secure development is an ever-evolving field that has advanced quickly in recent years with initiatives like Secure Development Lifecycle (SDLC), Development Security Operations (DevSecOps), and Model-Based Security Engineering (MBSE). Despite the persistence of the security and design communities to include security in the design process, significant security breaches continue to occur. Our work reviews existing literature to determine the current state of the research at the intersection of these design and cybersecurity fields and ultimately proposes an integrative and systematic approach for developers to generate design principles that incorporate traceable security. This approach integrates security regulations and design principles and activities, encouraging compliance and security considerations at the earliest stages of the design thinking process.

Large Language Models offer a novel approach with low barriers to entry to potentially improve knowledge transfer in product development. After identifying knowledge barriers from literature that are potentially addressable through LLM-based applications, we analyze two GDPR-compliant LLM applications - ChatGPT Enterprise and Langdock - examining their key features: assistants and chatbots for both, and prompt libraries and LLM-based file search for Langdock. Then, we evaluate each feature’s potential to mitigate each barrier. Our findings show that assistants and chatbots provide wide-ranging support across many barriers, whereas prompt libraries and file search deliver targeted solutions for a narrower set of specific challenges. Given the numerous influencing factors and the rapidly evolving field of LLMs, the study concludes with a research agenda to validate the theoretical findings.

Current quantitative methods for estimating product-related environmental emissions face limitations in supporting sustainable design, particularly in second-life product strategies. This paper highlights challenges in accurately assessing emissions and environmental impacts under existing regulations, which often fail to reward designs enabling circularity. Through examples of current practices, it underscores methodological ambiguities and regulatory gaps, proposing a research agenda for improved tools and frameworks. These advancements aim to better support the design, production, and certification of sustainable, second-life-ready solutions, fostering more effective environmental impact reduction. Additionally, the paper emphasizes the need for regulatory adaptation to incentivize circular design practices, ensuring a fair evaluation of products conceived for second-life applications

Generative Design (GD) tools can produce high-performing components with complex geometries that are challenging to conceive via traditional methods. While potentially disruptive, GD tools have yet to achieve widespread use in industry. One reason is that current GD tools are limited to manufacturing methods capable of producing intricate geometries that GD often creates such as 3D printing. To overcome this barrier, this paper quantifies the benefit of altering generatively designed parts to use standardized elements like wire stock and sheet metal via processes such as CNC bending and water jet cutting. Using a parametric cost model, we show that parts incorporating standard components can halve the unit price for production volumes of only 4 parts. Finite Element Analysis (FEA) reveals that replacing up to 60% of part volume has minimal impact on performance. Our findings highlight a gap and opportunity in existing GD research.

This paper explores how creative preservation, affected by a regulatory framework, unfolds in the design of complex systems. Based on a case study of the Boeing 737 aircraft, it focuses on the role of grandfather rights, as part of the regulatory framework of aircraft design, as a precursor for creative preservation. The paper analyzes three design decisions related tot the evolving Boeing 737 aircraft models over a period of six decades and highlight the changing logic of creative preservation in relation to technology maturity, increasing complexity of design decisions, and expanded stakeholder involvement. Overall, the paper demonstrates that the management of design heritage is a ‘living system’ and that foundational practices may slowly become ineffective.

Native to North America, Virginia pepperweed is a winter annual weed in the mustard family (Brassicaceae) found commonly in agricultural crops, roadsides, landscapes, and other undisturbed areas. Known for its peppery taste, Virginia pepperweed has emerged as a troublesome and difficult-to-control weed in and around major row crops in the Mississippi Delta region. Recently, Virginia pepperweed management has become increasingly challenging due to the weed’s ability to survive control measures when applications are made beyond its early rosette stage and high fecundity rates (∼100,000 seeds plant−1). Therefore there is a need to develop effective control measures that could reduce the spread of Virginia pepperweed in crop production systems. Greenhouse experiments were conducted in the 2024 season to evaluate the activity of various burndown herbicides labeled for Virginia pepperweed control in row crops. Virginia pepperweed seed was stratified and germinated in a growth chamber, and seedlings were transplanted into pots and kept in a greenhouse. The herbicides tested at the 1X rate were glyphosate at 1,261 g ai ha−1, glufosinate at 672 g ae ha−1, 2,4-D at 1,065 g ai ha−1, and paraquat at 840 g ai ha−1. Herbicides were sprayed at three growth stages: early rosette, late rosette, and bolting. Virginia pepperweed control was evaluated at 1, 2, 3, and 4 wk after herbicide application (WAA). Shoot dry biomass data were collected at 4 WAA. Application of 2,4-D resulted in 95% to 100% Virginia pepperweed control at all three growth stages. Depending on the growth stage at which herbicides were applied, there was 40% to 50% control with glyphosate, 18% to 47% with glufosinate, and 0% to 71% with paraquat, with 0% biomass reduction at the bolting stage. However, the highest dry biomass reduction (>80%) was observed with 2,4-D applications at the early rosette stage. Herbicide applications at the early rosette stage resulted in maximum Virginia pepperweed control.

Combinatorial optimization (CO) is essential for improving efficiency and performance in engineering applications. Traditional algorithms based on pure mathematical reasoning are limited and incapable to capture the contextual nuances for optimization. This study explores the potential of Large Language Models (LLMs) in solving engineering CO problems by leveraging their reasoning power and contextual knowledge. We propose a novel LLM-based framework that integrates network topology and contextual domain knowledge to optimize the sequencing of Design Structure Matrix (DSM) —a common CO problem. Our experiments on various DSM cases demonstrate that the proposed method achieves faster convergence and higher solution quality than benchmark methods. Moreover, results show that incorporating contextual domain knowledge significantly improves performance despite the choice of LLMs.

Prototyping is an important component of the engineering design process and has become a frequently studied topic in engineering education. The iterative strategy of creating prototypes, where a single design is refined with repeated improvements, is widely taught and considered to be the default approach to prototyping. However, research has shown that a parallel approach to prototyping, where multiple concepts are tested simultaneously, has potential benefits when exploring a complex design space. Recent studies on parallel prototyping in first-year engineering classrooms have shown that students required to use a parallel strategy produced higher performing final designs than students who used an iterative strategy. This work places the parallel and iterative prototyping strategies in a typical classroom setting where first-year engineering students have control over their strategy.