Machine learning approach for biological pattern based shell structures

Following previous research towards the subject of digital fabrication of thin shell structures, architectural generative design processes sharing similar physical and geometrical characteristics with biological processes were translated to fabrication processes, blurring the lines between physical, digital and biological, and allowed to examine the structural efficiency of segmented stripes arrangements of complex surfaces with less material usage. The goal of this paper is to examine the efficiency of implementing a machine learning approach into an already established design workflow and to develop a creative methodology for decision making. In order to specify the appropriate features, we look at related work that integrates machine learning inside the design and fabrication process.