IN-PERSON Lisbon Livable Cities. Section B

Abstract

This paper explores the potential of applying artificial intelligence (AI) and pattern languages to enhance infill development in historic places. We examine how this approach could democratize the design process, fostering more inclusive preservation practices through enhanced community engagement. As preservation efforts grapple with integrating new structures into established contexts, this research proposes a novel approach that harnesses AI’s predictive power alongside the timeless wisdom encapsulated in architectural pattern languages. We demonstrate how machine learning algorithms can help analyze existing historic fabric to identify and develop key spatial and architectural patterns that contribute to a location’s unique character; its “genius loci” or “deep place.” These patterns, which might include building height relationships, facade rhythms, and material textures, etc., form the basis for using AI-assisted design tools that can help generate context-sensitive infill proposals guided by traditional patterns. When tasked with bridging past and future, could this synthesis of AI and traditional wisdom be the key to more harmonious historic districts? While questions remain about AI’s ability to interpret subtle architectural nuances, we believe the potential benefits warrant further exploration. Our goal is to find a way to respect tradition while embracing new technology, helping to make historic places even better through thoughtful development of new buildings that fit in well. This research aims to bridge the gap between tradition and innovation, offering a path forward for enhancing historic places through thoughtful, AI-guided infill development. We invite discussion on this emerging approach and its implications for our shared future.