Article Preview
TopIntroduction
The Yao traditional buildings (YTB) in this study are located in the Guangxi Province in China and are a form of local indigenous architecture and represent a local cultural heritage. There are no specific records for the Yao ethnicity’s foundation due to numerous migrations at various scales, accompanied by regime changes throughout different dynasties in the Guangxi region. Timothy Oaks is a leading researcher into the origins of Yao ethnicity, and indicates that they took preliminary shape during the Song Dynasty (960‒1279) and became further established during the Yuan Dynasty (1286-1368) (Oaks, 1997).
The Yao buildings were and are still built by master carpenters and builders. Due to the exclusive use of wood material in the construction of these houses, many are no longer usable and some have been left unattended, demolished or replaced by contemporary structures in concrete. The focus of this body of work is concentrated in documenting the built forms of the houses and their spatial compositional principles. A systematic data collection in the form of 3D scanning and 3D modeling was undertaken. More than twenty existing buildings in Liaojia village in Guangxi have been identified for further survey, and three of them have been surveyed for this research project. It is the objective of the research outlined in this paper to bring a new dimension to the study of YTB by employing a formal approach to simulate and capture the architectural diversity of YTB buildings in a computational process. The approach includes firstly identifying the common structure and appearance of YTB, secondly developing the genetic-generative algorithm to construct multiple instances of the Yao style buildings, and lastly providing a computational model to digitally rebuild the YTB diversity, contributing in this way to preserving the memory and architectural heritage of YTB in Guangxi Province of China. This article is part of a research project intended to capture the uniqueness of each YTB, and to store and display in a digital format endless possible combinations of building structures. The compositional rules to build the generative computational system were developed by referring to the construction methods described in previous documentations on YTB and other wood constructions characteristic of other Chinese ethnic groups, and through direct survey of three selected case studies through 3D scanning techniques. The end result of this study is a modeling and generation tool that can recreate a particular recorded YTB or generate a new YTB that conforms to this particular architectural style.
Computer simulation of evolutionary processes is a well-established technique for the study of biological dynamics. Currently there is the possibility of generating a population of virtual plants or animals in a digital environment and analyzing how they transform as they pass on their virtual genetic material to their progeny. Aristid Lyndenmayer (1968) was one of the pioneers in computationally developing three-dimensional models of plants based on principles of genetics - a system originally created for the modelling of filamentous organisms and plant like structures. These are, in turn developed and progressed using the evolutionary computation technique of Genetic Algorithms (GA’s). GA’s are employed for their ability to assign adaptability values to each new form. Determining how a gene spreads in a population along the way - over several generations - is a task that some computer programs, collectively known as GA’s, carry out automatically.
The study of the formal and functional properties of this type of program has today become a field of knowledge in its own right, with an increasing applicability in architectural design. This article does not deal with the computer aspects of genetic algorithms (as a special category of “search algorithms” but with their application in the field of design, in specific, with the applications that this technique can have as an auxiliary tool for rebuilding digitally the YTB.