Yeomans et al. [37] combine a genetic algorithm with simulation
to solve the problem of municipal waste flow allocation under
uncertainty, improving the work in Huang et al. [36]. In particular,
each candidate solution of the population set, which contains
uncertain elements such as the quantity of waste generated at
each district, is simulated to be evaluated. Then, based on the
results of the simulation phase, the genetic algorithm allows the
system to evolve toward better solutions, generating a new set of
candidates to be evaluated again by simulation. After termination,
the algorithm provides, in addition to the best solution, a number
of “good” solutions. In this way, the method could be used even for
policy comparisons purposes. The procedure is tested on real data
from the same case study as in Huang et al. [36]. When compared
to the results of such study, the proposed approach is able to
improve over them, with gains ranging from 4% to 6%.