Traveling Salesman

TravelingSalesmanProblem

Description

A data structure to hold the problem definition of the traveling salesman problem.

Fields

• distancematrix::Matrix{Real}: The distance matrix representing the distances between cities.

solve(problem::TravelingSalesmanProblem; popsize = 100, ngen = 1000, pcross = 0.8, pmutate = 0.01, nelites = 1)::TravelingSalesmanResult

Description

Given a matrix of distances wrapped in a TravelingSalesmanProblem, returns a TravelingSalesmanResult with the best route and its cost.

Arguments

• problem::TravelingSalesmanProblem: a TravelingSalesmanProblem instance containing the distance matrix
• popsize::Int: the population size. Default is 100
• ngen::Int: the number of generations. Default is 1000
• pcross::Float64: the crossover probability. Default is 0.8
• pmutate::Float64: the mutation probability. Default is 0.01
• nelites::Int: the number of elites. Default is 1

Returns

• TravelingSalesmanResult: a custom data type that holds the best route and its cost

Example

pts = Float64[
    0 0;
    0 1;
    0 2;
    1 2;
    2 2;
    3 2;
    4 2; 
    5 2;
    5 1;
    5 0;
    4 0;
    3 0;
    2 0;
    1 0;
]

n = size(pts, 1
distmat = zeros(n, n)

for i in 1:n
    for j in 1:n
        distmat[i, j] = sqrt(sum((pts[i, :] .- pts[j, :]).^2))
    end 
end

result = solve(TravelingSalesmanProblem(distmat), ngen = 1000, popsize = 100, pcross = 1.0, pmutate = 0.10)

TravelingSalesmanResult

Description

A data structure to hold the result of the traveling salesman problem.

Fields

• route::Vector{Int}: The best route found.
• cost::Float64: The cost of the best route.