**6.3 Conclusions on mathematical modeling of vegetables and fruits**

Mathematical modeling was performed based on the interpolation of experimental data obtained in this study.

	- the number of plants corresponding to the implementation of the phytoremediation process of soils contaminated with heavy metals;
	- the compliance of plant production within the limits provided for human consumption;
	- the optimization of phytoremediation and crops with regard to the marketing and consumption.

Statistical mathematical models determined with equations (Eqs.(14), (15) and Eq. (17)) whose coefficients are given in the **Tables 7**–**9**, for each case, they can be used for interpolation in the calculation of concentrations of heavy metals in the plant or calculation of transfer coefficients, only for the experimental range used in this study, for the initial concentration of each heavy metal in the soil.

• Mathematical modeling for fruits is poorly documented because at certain concentrations, plants have not developed. It is recommended for the two species of studied berries (strawberries and raspberries), interpolation curves of maximum second degree, because the experimental distributions are not linear. Interpolation was also attempted using third degree polynomial curves, which passed through experimental points but had a total zero error.


• It was observed that the variation of the heavy metal content in berries (fruit) does not recommend linear interpolation (linear regression), as these distributions have potential extreme points (minimum or maximum).
