Solving for Shape with PDEs in Geometric Modelling

It is tempting to think of shape not simply as something we draw, but as something we can solve for. Partial differential equations, particularly the biharmonic equation, offer a way of generating smooth surfaces from a relatively small amount of geometric information. Instead of specifying every point on an object, we can define its boundaries and constraints, then allow the equation to determine the surface that lies between them.

This is especially attractive for complex or organic forms, where traditional modelling can require many separate patches and a great deal of manual adjustment. A mathematical surface can respond smoothly when its boundaries are changed, making it possible to create and modify forms such as aircraft, vessels, animals or other free-form objects in a more unified way.

The broader idea is that mathematics might become part of the design language itself. Rather than treating geometry as a fixed collection of polygons, we can begin to treat it as the outcome of an underlying physical or mathematical process—potentially giving us models that are smoother, more compact and easier to manipulate.