This page contains the HGA code proposed in the following paper:

Mahmoudinazlou, S., & Kwon, C. (2024). A hybrid genetic algorithm for the min–max Multiple Traveling Salesman Problem. Computers & Operations Research, 162, 106455.

How to Install as a Julia package?

cd m-TSP
julia
] add .

Test:

using HGSmTSP

n_vehicles = 3
n_nodes = 10
coordinates = rand(n_nodes, 2) .* 1000

dist_mtx = zeros(n_nodes, n_nodes)
for i=1:n_nodes
    for j=1:n_nodes
        dist_mtx[i, j] = sum(abs.(coordinates[i, :] - coordinates[j, :]))
    end
end

@time solve_mTSP(n_vehicles, dist_mtx, coordinates)

In the above dist_mtx is a matrix of size n+1 by n+1 where n is the number of customers. The first node is the depot, and there is no dummy node added. coordinates is n+1 by 2 matrix.