USI AI Cup - TSP Solver

My solution to solve the Traveling Salesman Problem (TSP) as part of the AI course at the USI.

Table of Contents - cause is cool to have

• Algorithm
• My Results
• Conclusion

Algorithm:

The algorithm used in the code starts with an initial solution generated using the nearest neighbor heuristic. It then employs the 2-opt algorithm to improve this solutionthe simulated annealing as optimization method. The hope is that simulated annealing will allow the algorithm to explore a broader range of solutions and find a tour close to the optimal.

My Results:

Problem	Best Known	Estimated	Error
ch130	6110	6110	0,00%
d198	15780	15852	0,46%
eil76	538	541	0,56%
fl1577	22249	23224	4,38%
kroa100	21282	21343	0,29%
lin318	42029	43600	3,74%
pcb442	50778	52904	4,19%
pr439	107217	112825	5,23%
rat783	8806	9308	5,70%
u1060	224094	241583	7,80%
FINAL RESULT			3%

Conclusion:

Go check Ant Colony Optimization instead, is better