The optimization in design problems is elusive due to the inherent multiplicity and ambiguity of the optimal for real world problems. However partial optimization of specific aspects of architecture is possible and can be implemented in both physical models [in the way hanging chains, membranes and stress induced birefringence has been used in the past] and digital simulations. The students will be required to develop one such physical or digital model. These models are only as good as the correspondence of their apparatus [physical or simulated] to the real world the often larger scale system they simulate. Hence the important when designing optimization apparatuses is not the algorithm used but the robustness of the system and the understanding of its limitations on the part of the designer. Students can choose to engage with existing simulation and analysis software, develop their own or use physical models for optimization and any combination of the above.