DesignFusion: Integrating Generative Models for Conceptual Design Enrichment
Liuqing Chen,Qianzhi Jing,Yixin Tsang,Qianyi Wang,Lingyun Sun,Jianxi Luo
DOI: https://doi.org/10.1115/1.4065487
2024-01-01
Journal of Mechanical Design
Abstract:Abstract Conceptual design is a pivotal phase of product design and development, encompassing user requirement exploration and informed solution generation. Recent generative models with their powerful content generation capabilities have been applied to conceptual design to support designers' ideation. However, the lack of transparency in their generation process and the shallow nature of their generated solutions constrain their performance in complex conceptual design tasks. In this study, we first introduce a conceptual design generation approach that combines generative models with classic design theory. This approach decomposes the conceptual design task based on design process and design attributes, and uses the Who, What, Where, When, Why, How (5W1H) method, Function-Behavior-Structure (FBS) model, and Kansei Engineering to guide generative models to generate conceptual design solutions through multi-step reasoning. Then we present an interactive system using a mind-map layout to visualize multi-step reasoning, called DesignFusion. This empowers designers to track the generation process and control inputs/outputs at each reasoning step. Two user studies show that our approach significantly enhances the quality of generated design solutions and enriches designer experience in human-AI co-creation.