Conference Agenda

Despite the rapid advancement of generative Large Language Models (LLMs), there is still limited understanding of their potential impacts on engineering design (ED). This study fills this gap by collecting the tasks LLMs can perform within ED, using a Natural Language Processing analysis of 15,355 ED research papers. The results lead to a framework of LLM tasks in design, classifying them for different functions of LLMs and ED phases. Our findings illuminate the opportunities and risks of using LLMs for design, offering a foundation for future research and application in this domain.

This study explores the application of image generative AI to support design process by creating inspiration boards. Through an evaluative study, we compare the diversity, quantity, fidelity, and ambiguity of boards generated by image generative AI and traditional methods. The results highlight how generative AI produces a quantity of images, it exhibits limited diversity compared to traditional methods. This suggests a tendency for supporting interpolation rather than extrapolation of ideas, in turn providing insights on best practice and into the optimal stage for its application.

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 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.

Sketch2Prototype is an AI-based framework that transforms a hand-drawn sketch into a diverse set of 2D images and 3D prototypes through sketch-to-text, text-to-image, and image-to-3D stages. This framework, shown across various sketches, rapidly generates text, image, and 3D modalities for enhanced early-stage design exploration. We show that using text as an intermediate modality outperforms direct sketch-to-3D baselines for generating diverse and manufacturable 3D models. We find limitations in current image-to-3D techniques, while noting the value of the text modality for user-feedback.

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.