Topology optimization refers to the practice of using an iterative, software-driven process to optimize the distribution of part material within a given 3D space. It can also be considered as the computational extension of the philosophy of product design. In topology optimization, stress analysis is applied to existing part geometry. Software inputs include the location and direction of forces and stresses, fixed surfaces/mating points, a fixed design space (which may be larger than the actual part geometry), selected material and its properties, and the desired outcome.

Generative design is also like topology optimization, except original part geometry is not provided. Instead, the final part function is defined within a given design space according to similar instructions as with topology optimization (e.g. stresses, fixed locations etc.). The software then iteratively produces a series of solutions that achieve the desired performance criteria of the final part within the design space.

Generally, both approaches will create a design that is non-optimal for manufacturing, because the material is not allocated against a condition of manufacturability but instead the final part function. In addition, because each approach typically categorises the part geometry into a series of volume elements, the result may appear coarse hence Clean-up is needed for smoothing of imperfections.

Topology optimization interpreted as sacrifices in the design must be made for manufacturability. The principle is that the topology optimized shape can be maintained, and the final weight and structural properties can be closer to that of the optimized shape. Reducing the weight also means that the manufacturing costs would be less.