A grammar-based optimization approach for walkable urban fabrics considering pedestrian accessibility and infrastructure cost

Designing urban areas that provide smaller distances to their amenities is a key factor toward more walkable environments. Moreover, this is a critical aspect of climate-resilient urban planning since it is broadly assumed that areas with greater walkability discourage automobile usage and reduce CO₂ emissions. Generative and data-driven design approaches, in turn, increase designers’ ability to explore wider sets of potential solutions. In this sense, identifying designs with an optimized performance out of the vast possibilities that computation can provide is crucial. Shape grammars are a formal method of shape generation that facilitate the elaboration of complex patterns and meaningful designs. This paper hypothesizes that coupling shape grammars with multi-objective optimization can help address trade-offs and decision-making in urban design. It focuses on the pedestrian accessibility and infrastructure cost (as estimated by cumulative street length) trade-off in urban fabrics as a case study to verify the suitability of a grammar-based optimization approach for more dynamic and efficient solution-finding in urban design. Our findings suggest that a grammar-based optimization approach is helpful in addressing urban trade-offs as it could be used to filter the design space and provide optimal alternative fabric layouts with increased pedestrian accessibility and decreased infrastructure cost.