Generative design of floor plans of multi-unit residential buildings based on consumer satisfaction and energy performance

Abstract

Satisfying consumers is an important task in the design of residential floor plans. However, the design of residential floor plans with the aim of improving consumer satisfaction and energy performance has not been fully explored. Also, the preferences of different customer groups regarding customer satisfaction have not been studied. Traditional design and optimization methods still have limitations in terms of performance improvement, while performance-based generative design methods can explore the performance improvement potential of a larger range of solution spaces. Therefore, this study introduces a generative design framework that is grounded on optimizing energy performance and consumer satisfaction for multi-unit residential floor plans, which consists of four main processes: 1) Generation of floor plan schemes. 2) Evaluation of consumer satisfaction and energy performance of the generated schemes. 3) Running a multi-objective optimization to obtain optimal schemes. 4) The final floor plan scheme is determined by the decision maker. The proposed framework was employed in a real design case, and seven optimal solutions were obtained. Compared with the initial scheme, the maximum improvement in consumer satisfaction and energy performance was up to 38.78% and 11.18% respectively. The results also indicated that the preferences of different types of consumers differed significantly. Thus, different evaluation functions have been created to provide solutions with higher satisfaction levels targeting different types of consumers. As such, generative design and optimization of residential floor plans for specific types of consumers can be undertaken at an early design stage to improve consumer satisfaction and reduce energy consumption.

Keywords

• Performance-based generative design
• Residential floor plan design
• Consumer satisfaction
• Energy performance