Implementation of the Simulated Annealing Algorithm in solving the (Symmetric) Travelling Salesman Problem.
In the outer loop of our algorithm, we reduce T by setting T = 0.9 * T. While, in the inner loop, we obtain L different solutions at each T. These solutions are generated by randomly swapping two indexs(cities) of tour x.
Then we apply the Metropolis acceptance Probability=min {1, exp(-(fy-fx)/T)} as the probability of accepting a new possible solution. The algorithm then repeats while T > ε, with ε = 10^{-2}, and stops otherwise.