Design Concept Generation

Design concept generation aims to automate the creative process of developing new ideas, traditionally reliant on human brainstorming and intuition. Current research heavily utilizes large language models (LLMs), particularly generative pre-trained transformers (GPTs), to generate design concepts from textual prompts, exploring various approaches like bio-inspired design and different types of creative reasoning (e.g., analogy-driven, problem-driven). These methods are evaluated using both human expert assessments and computational metrics, focusing on improving the feasibility, usefulness, and novelty of generated concepts. This work holds significant potential for accelerating the design process across various fields by providing designers with novel and useful starting points for innovation.