Many flow problems solved in engineering practice involve complex geometries. The creation of structured or block-structured meshes (consisting of quadrilateral or hexahedral elements) for such problems can be extremely time-consuming, if not impossible. Setup time for complex geometries is, therefore, the major motivation for using unstructured meshes employing triangular or tetrahedral elements. If your geometry is relatively simple, however, there may be no clear saving in setup time with either approach.

For many typical geometries, the use of triangular or tetrahedral elements leads to a significantly higher number of elements than for quadrilateral or hexahedral elements to obtain similar accuracy on the same geometry. This higher number of elements results in a higher CPU cost. In this respect, a mesh made of triangles or tetrahedra is often less than optimal in terms of CPU cost.

The most efficient mesh is one that includes quadrilaterals or hexahedra in regions where the geometry is not too complex, and triangles or tetrahedra in regions where it is difficult to place quadrilateral or hexahedral elements.

Some 3D remeshing techniques such as Optimesh, streamwise, and planar require the nodes to be defined within a plane. In addition, due to the nature of the remeshing technique, the node will be relocated in the plane orthogonal to the extrusion direction only. Therefore, such remeshing techniques require the use of brick (hexahedral) or prism elements in the extrudate (although tetrahedra and pyramids can be used in the region that is not being remeshed).