Automated Layout Zoning : A Case of the Campus Design Problem

Abstract

Layout zoning is the foremost and crucial step in the design of large-scale greenfield construction projects. While it is known that early design decisions have significant impact on the social and environmental value of the completed facility, much of the relationship between design attributes and values are tacit knowledge. In addition to the challenges, such as the number of design alternatives that can be evaluated in a limited time, traditional methods of design generation and evaluation do not focus on capture of stakeholder defined values and their quantification for effective evaluation of design alternatives. Computational design tools provide several benefits that can be leveraged in automated design generation to overcome the limitations of conventional methods of design. However, the translation of stakeholder defined value to parametric form in automated design generation is not adequately explored in existing works. In this work, generation and evaluation is formulated in a visual programming environment -Grasshopper. The ‘growth algorithm’ discretizes the available area into numerous land parcels and allocates the parcels to seed nodes until the area requirement is satisfied. The implementation in the case of a real-world campus layout revealed that the script was able to generate layout alternatives with different configurations of zones. Further, from the case study, a set of values and their relationship with design attributes were derived. Using parametric scripting, it was illustrated how the value-based design objectives can be quantified by utilizing the layout solutions produced by the growth algorithm-based script.