**5. Conclusions and impact of the research**

From the civil engineering point of view and regarding the laboratory tests, it must be established that the low walls supporting the slab bent within the permissible limits since there were already headframes confining them. On the other hand, there was no reinforcing steel in the temperature nerve, so it did not function as such and that was a noticeable aspect. In addition, the surface of the slab on which the smooth steel rod was placed was not smooth enough, that is, the surface of the slab was rugose, which caused the load again not to be completely vertical, but rather, that the load might have had a certain unknown tilt angle but is supposed to be lower than the angle generated in the first test, so that applied load may be diminished in a certain percentage lower than the previous one. Taking into account the deformations obtained in this test, they were also higher than the permissible ones, above all, in the second and third loads. In the first test, it was almost the permissible one: 2.61, 4.73, and 5.98 cm, respectively.

The element worked satisfactorily, and while it is true that its deformations were not permissible, this was because the loads applied are not the ones usually received by a slab, and in cases where it is necessary to support a load of such magnitude, it has to be replaced with an element with higher load capacity, such as a slab or if necessary a column, **Figure 8**.

Regarding the fractal part, the methods used in this work are related to the fractal line or curve taken as a unit and the number of segments into which it is divided and the fractal dimension is always between one and two, but not higher than two. Logarithmic units are used, since smaller scales of the unit are progressively used, which are proportionally mirrored, that is why the logarithm is used, due to the smaller proportionate scales that appear. In the equations of straight lines shown in **Table 1**, we may see that the exponents to which the variable *x* is raised are almost identical, except for the one obtained through the ruler-dimension method.

Whether using the box-dimension method, the ruler-dimension method, or the mass-dimension method, these methods are related to the logarithm of *N* (which is the unit or the whole of a segment) between the numbers of *s* (which is the number of parts into which *N* is divided).

The increase of regression lines has also been observed, but they do not change since they follow the same increase direction. This means that the fracture never gets to failure, which will increase infinitely at higher scales, and which, to the eye, one might think is a curve, but from a closer point of view, one would realize that there are the same fractal lines but very small. These are imperceptible to the human eye and are only visible through other instruments or programs that may increase the scale of

**Figure 8.** *Large deformation and cracking a short time before the collapse.*

#### **Figure 9.**

the drawing. In addition, it has been observed that *Status 3*, which is still the fractal axis but at a higher scale, shows us that it is very similar to the borders of both ends, which at a higher scale would be more similar to the fractal axis, with a higher *status n* on the borders of the fracture. This shows that no matter the distance of the opening of the fracture, it will always be the same and/or similar to the fractal axis on both ends, **Figure 9**.

Regarding the theoretical part related to the fractal intersection, it has been proven that if there are two collinear fractal curves, their intersection is a segment of a line. If they are cut with a certain angle, **Figure 2**, such an intersection is a singular point and the rigid movements may be parameterized through coordinates (*x*, *y*, θ); with θ being the rotation angle. In addition, the fractal geometry obtained with the fractal in **Figure 3** is impressive upon considering the large intersections between them.
