Work

Algorithmic Graph Grammar for Generative Design of Complex Systems

Year: 2026
Type: preprint
Abstract: This paper presents a novel algorithmic approach to generative design, leveraging graph grammars to create complex system designs. Traditional generative design methods frequently struggle with the inherent complexity of systems, often resulting in designs lacking creativity and efficiency. Our core innovation lies in the development of a graph grammar-based system that allows for the automated creation of intricate structures – encompassing networks, architectural designs, and workflows – by systematically defining and applying transformation rules. These rules are driven by both design constraints and aesthetic principles, providing a flexible framework for exploring diverse design solutions. The system's architecture is detailed, including the input representation, rule definition, execution engine, and output format. We demonstrate the applicability of this approach through theoretical examples and outline a potential implementation strategy, highlighting its advantages over conventional generative design techniques. The key contribution is a formalized methodology for generating complex systems that embraces both structural logic and design intent.
Language: English
Subfield: Architecture
Open Access status: green
OpenAlex ID: W7204863592
Cites:
Cited by:
Related to: 10