During the development of ShallowCore, the results obtained for a number of typical and extreme examples were compared with calculations performed using software commonly used in design practice and manual calculations.

In case of a difference in the intermediate or final result greater than 1%, the cause of the difference was thoroughly investigated. The interpretation of NEN9997-1 was carefully considered and during the development and validation process, special attention was paid to:

- Determination of the zone of influence for undrained calculation.
- Treatment of thin layers immediately below foundation level.
- Shallow foundations near embankments.
- Calculating different scenarios for a design situation.
- Applying reinforcement for increasing the bearing capacity.

As an example of the validation, significant differences were found between ShallowCore and software commonly used in design practice for situations where an embankment is present in the area of influence of the foundation. Two explanations were found underlying this discrepancy:

- Interpretation of table 6.a/figure 6.c from NEN9997-1 for the width of the zone of influence in the undrained scenario.
- Differences in the calculated angle ß and its effect on the correction factors due to the ground level slope (λ) .

The first discrepancy is discussed in more detail below.

For the interpretation of table 6.a/figure 6.c from NEN9997-1, it is concluded that the choice of a significant influence on the calculated width of the zone of influence (ae), especially for characteristic values of the friction angle in the range 0° ≤ ϕ'k ≤ 5°. The graph below shows the divergence of appropriate polynomials of different degrees in the lower range of friction angle values.

Figure 6.c shows the relationship between the influence width (ae) and the effective width of the foundation (b') depending on the characteristic value of the angle of internal friction (ϕ'k) and the ratio of horizontal to vertical load (H/V). For the undrained situation, ϕ'k = 0o and therefore is the ratio ae/b' equal to the intersection with the y-axis (ϕ0). For Figure 6.c, and also Table 6.a (NEN9997-1), the ratio ae/b' is only given for ϕ'k ≥ 5o and thus not for the ϕ'k = 0o. Thus, to determine the influence width for the undrained situation, extrapolation is required. Here the result depends on the degree of the polynomial used for the extrapolation. For H/V = 0, ae/b' varies between 0.34 (n=3) ≤ ϕ0 ≤ 2.28 (n=2).

For the second-degree polynomial, the value for ae/b' is above the value as found for ϕ'k = 5o. Given the shape of the lines and the numerical value in Table 6.a, the ratio ae/b' is expected to decrease at values of ϕ'k < 5o. Therefore, based on Table 6.a from NEN9997-1, for ϕ'k < 5o and H/V = 0, an intersection point smaller than 1.25 can be expected. Differences in extrapolation may lead to significant differences in the results of the bearing capacity calculations. Here, differences up to more than 23% are possible.

For the full discussion with examples and numerical values, please view our ShallowCore validation report. Bear in mind that it is written in Dutch.