Implementation and Evaluation of Generative Layout Options Using the Graph Theoretical Approach for a Hospital Layout Problem

Abstract

The effective layout of departments within a new hospital influences the efficiency and effectiveness of delivering healthcare services. Given a set of healthcare departments, adjacency ratings, areas, and structural bay sizes, a graph theoretical approach (GTA) was used with a novel placement strategy to generate an initial set of optimal and near-optimal layouts based on a distance-weighted adjacency score. Healthcare planning and design experts were surveyed to compare their results to the layout scores. This study explores a computational methodology for automatically generating many near-optimal layout options in a new hospital facility, provides a methodology to generate layout massing and connect that data to building information modeling (BIM) authoring programs, and provides insight into experts’ perceptions of using an automated approach to support layout design. The main findings from this study indicate that: (1) experts’ scoring of layouts aligns well with the scoring metric, (2) experts found generative layout methods to be beneficial to explore more options than traditionally possible and alleviate design bias, and (3) future method development should make decision information available to users.