The cost values in Table 6 are obviously dependent on the
assumptions made during the cost estimate. Preliminary estimates
for the equipment cost are affected by uncertainties, while the purchase cost of reactants can fluctuate both geographically and
temporally. Furthermore, even the management costs for solid
residues might vary with time: for instance, additional taxes on
landfill disposal could be raised by policymakers in order to better
compensate the externalities generated by the environmental burden
of landfilling, whereas costs related to the brine recovery process
might drop thanks to technological improvements.
Therefore, a sensitivity analysis was performed in order to
check the robustness of the results even for widely different values
of the unit cost entries. A Monte Carlo approach was adopted. The
total annual cost was expressed as a function of 6 stochastic variables:
the total installed costs of the ESP and the FF device, the purchase
costs of lime and bicarbonate, the disposal cost of residues
and the cost of recovering brine from SBW. A typical distribution
of probability (a symmetric beta distribution with a = b = 2) was
associated to each variable, within the respective intervals of variation
reported in Table 7. The mean values were set equal to the
base values discussed above and used in the assessment, while
ranges were selected according to the scattering of cost data discussed
in Supplementary Material. The largest range of variability
was given to the disposal cost, in order to reflect the uncertainty
due to the wide variety of alternatives (stabilisation and disposal
in common landfill or disposal in hazardous waste landfill or backfilling
in salt mines) and the non-homogeneous regulatory framework
even within the EU.