I recently wrote an article about Sean Hanna’s and Siavash H. Mahdavi’s research into microstructure optimization for Architektura – Murator – a polish architectural design magazine.

Sean Hanna is a RCUK Academic Fellow in Space and Adaptive Architectures based at the Bartlett School of Graduate Studies of UCL. His microstructure optimization is essentially a design of an optimized material to be produced with SLS. It is conceptually grounded by an observation of the behavior of the cells of wood, which represent a degree of local intelligence. They adapt (optimize) their form according to their position in the overall structure. Microstructure is also composed out of cells, which are optimized in two ways (or for two kinds of conditions).

optimization for manufacturing constraints

First kind of optimization is an multiple-objective optimization for the digital manufacturing technique constraints. A genetic algorithm is used in order to optimize the shape of the elements composing each cell. The goal is to be efficient in terms of the amount of material used but still be able to produce the object with an SLS machine. Underneath you can see an example of a population of solutions.

structural optimization

Second type of optimization is structural optimization. Forces in a specific shape which is about to be produced in microstructure are analyzed. Then the thicknesses of the cells are adjusted accordingly allowing for even greater efficiency of the material. Support vector machines are used in this process. In opposition to the optimization for manufacturing constraints, this type of optimization has to take place every time new object is about to be produced out of microstructure.

sample – a cantilever

A cantilever was chosen as a sample of microstructure. The deflection of the microstructure cantilever is several times smaller than the one of a standard engineered structure (image above). The sample has only 2 cm but I believe the potential of the technique is enormous.

images: courtesy of Sean Hanna