Sahoo et al. [44] discuss how to divide the area from which
waste is to be collected into districts in order to subdivide the
problem and make it more manageable. In particular, they propose
a mathematical model, as well as a two phase insertion algorithm,
in which a feasible solution is first generated, and then improved.
The first phase of the solution approach utilizes a k-meansvariant-
balanced-clustering algorithm [45], that randomly selects
initial centroids, and then clusters the stops according to the
distances between the stops and the centroids. The second phase
makes use of an extended version of the insertion algorithm by
Solomon [46], and a simulated-annealing metaheuristic combined
with the CROSS exchange local-search method of Taillard et al. [47]
for further improvements.