A particular knowledge and application gap in Interactive Genetic Algorithms (IGAs) is associated with their capacity to accommodate human intuition or subjectivity alongside optimization processes. This is especially important for architectural design, where subjective and objective reasoning are employed. In response, this research has three aims. First, a comprehensive framework for identifying types of interaction in IGAs must be developed. Second, to understand whether current IGA tools can accommodate the required levels and modes of subjective interaction in architecture. Third, to propose a new IGA-based tool for architecture that addresses both the knowledge and application gaps identified in the first two aims. This new tool, Snowflake, enhances interactivity and guides the process towards optimized solutions while integrating architects’ subjective preferences.
