Mathematically derived minimal surfaces offer great potential for large-scale explorations due to their minimised area and weight. This thesis develops applications for minimal surfaces by exploring its geometric properties. These are cost-effective solutions, such as mold re-use for fabricating panels. Here, minimal surfaces are generated using a constraint-based solver. A 2D matrix is used to develop a k-means clustering algorithm that results in grouping panels who share curvature parameters irrespective of their location in the mesh.

Results show clustering roughly follows the isolines of the principal curvatures of the mesh. This is due to minimal surfaces being a special type of Weingarten surfaces and therefore have a functional relation between its two principal curvatures. An attempt is made to discretise the geometry following the curvature isolines to examine the behaviour of the resulting panels. The development of the computational tool set can be used to support the design of panelisation of minimal surfaces. It can provide designers with information about ideal panelisation or indication of performances of custom made panelisation suggested by the designer.