Computational Design for Sport Building

The design of sport buildings has great impact on top-sport as well as on recreational sport-activities. It implies challenging tasks in meeting the performance-requirements. This includes the control of factors like daylight/lighting, air flow, thermal conditions, just to name a few. Such factors impact the performance of athletes and are hard to control in large sport halls; their control is even harder when the public/audience is located within the halls and require different climate conditions. While mechanical installations are often needed during competitions in order to guarantee constant conditions, relaying on mechanical installations during the daily and recreational use of the venues challenges their medium/long term sustainability. Computational form finding approaches can favour the achievement of high-performing and sustainable sport buildings. In this light, the paper tackles the use of Multi-objective and Multidisciplinary design optimization. The paper presents the concept of Multi-objective Multidisciplinary design optimization techniques to support trade-off decisions between multiple conflicting design objectives and interdisciplinary design methodology, during the conceptual design of sport buildings. The proposed method is based on parametric modelling, performance simulation tools and algorithms for computational optimization, for which the paper tackles three specific aspects. First of all, due to the complexity of large sport buildings, the formulation of the optimization and the screening of the related design variables is crucial in order to obtain a meaningful design space, which helps reducing unnecessary computational burden. Secondly, assessing performance based on measurements and analyses is crucial and can be supported by performance simulations tools; however effectively integrating performance simulations tools in the early phase of the design requires new tools. In this light, a customized computational process for the rapid assessment of temperature and airflow patterns is presented. Thirdly, the process requires the combination of design optimization and design exploration, while searching for well-performing solutions. The importance of design exploration is emphasized also for sub-optimal solutions. In order to facilitate the design exploration, the combination of optimization algorithms, multi-variate analysis algorithms and options for exploring design solutions via an interactive dashboard connected to a database are presented. To exemplify the method, specific case studies are developed as collaboration between Delft university of Technology and South China university of Technology.