One has

Denoting 
 the jacobian

one finally has

When 
, the expression of the jacobian reduces to 
.
With 
 a part of the boundary of 
 a real element and
 the corresponding boundary on the reference element 
,
one has

where 
 is the unit normal to 
 on 
. In a same
way

For 
 the unit normal to 
 on 
.
Denoting
![\nabla^2 f =
\left[\frac{\partial^2 f}{\partial x_i \partial x_j}\right]_{ij},](../_images/math/922650db95a714a428af3c384a4e359d258de17e.png)
the 
 matrix and

the 
 matrix, then

and thus

In order to have uniform methods for the computation of elementary matrices, the
Hessian is computed as a column vector 
 whose components are
. Then, with
 the 
 matrix defined as
![\left[B_2(\widehat{x})\right]_{ij} =
\sum_{k = 0}^{N-1}
\frac{\partial^2 \tau_k(\widehat{x})}{\partial \widehat{x}_{i / P} \partial \widehat{x}_{i\mbox{ mod }P}}
B_{kj}(\widehat{x}),](../_images/math/af0dc2d4ddff4d7e7c23ea66da969f7cd3547afb.png)
and 
 the 
 matrix defined as
![\left[B_3(\widehat{x})\right]_{ij} =
B_{i / N, j / P}(\widehat{x}) B_{i\mbox{ mod }N, j\mbox{ mod }P}(\widehat{x}),](../_images/math/a154bfb20b9d7666d4d1ac1dc4f0ea257870ff0d.png)
one has

Assume one needs to compute the elementary “matrix”:

The computations to be made on the reference elements are

and

Those two tensor can be computed once on the whole reference element if the
geometric transformation is linear (because 
 is constant). If the
geometric transformation is non-linear, what has to be stored is the value on
each integration point. To compute the integral on the real element a certain
number of reductions have to be made:
) nothing.
) a
reduction with respect to 
 with the matrix 
.
)` a reduction of 
 with respect to 
with the matrix 
 and a reduction of 
 with respect also
to 
 with the matrix 
The reductions are to be made on each integration point if the geometric transformation is non-linear. Once those reductions are done, an addition of all the tensor resulting of those reductions is made (with a factor equal to the load of each integration point if the geometric transformation is non-linear).
If the finite element is non-
-equivalent, a supplementary reduction of the
resulting tensor with the matrix 
 has to be made.