A three dimension object is expressed mostly using a mathematical model; so, it is topical the question how we will represent it? Essentially, for this we have two methods:

First is the representation by a volume model where the object has a parametric representation. This method is suitable for subsequent processing and arrangements of a given object. It is used mainly in mechanical engineering production because it enables the subsequent physical simulation of the product features. A disadvantage of this method is that model designing is limited only for the use of basic geometric forms. 

The second method is area representation. Here we do not depend on geometric forms because the object is constructed only using polygons of its surface. The surface of an object comprises vertexes (points) interconnected by abscissas into polygons. In this way we can represent practically any object using a polygonal network, which from the point of view of universality is almost ideal. The most advantageous representation is if there are given polygons - triangles because calculations with these are the simplest.