Automatic generative design and optimization of hospital building layouts in consideration of public health emergency

Purpose: This study aimed to propose a performance-oriented approach of automatically generative design and optimization of hospital building layouts in consideration of public health emergency, which intended to conduct reasonable layout design of hospital building to meet different performance requirements for both high efficiency during normal periods and low risk in the pandemic. Design/methodology/approach: The research design follows a sequential mixed methodology. First, key points and parameters of hospital building layout design (HBLD) are analyzed. Then, to meet the requirements of high efficiency and low risk, adjacent preference score and infection risk coefficient are constructed as constraints. On this basis, automatic generative design is conducted to generate building layout schemes. Finally, multi-objective deviation analysis is carried out to obtain the optimal scheme of hospital building layouts. Findings: Automatic generative design of building layouts that integrates adjacent preferences and infection risks enables hospitals to achieve rapid transitions between normal (high efficiency) and pandemic (low risk) periods, which can effectively respond to public health emergencies. The proposed approach has been verified in an actual project, which can help systematically explore the solution for better decision-making. Research limitations/implications: The form of building layouts is limited to rectangles, and future work can explore conducting irregular layouts into optimization for the framework of generative design. Originality/value: The contribution of this paper is the developed approach that can quickly and effectively generate more hospital layout alternatives satisfying high operational efficiency and low infection risk by formulating space design rules, which is of great significance in response to public health emergency.