This project solves the Traveling Salesman Problem with metaheuristic Iterated Local Search (ILS) and compares the results with sequential and parallel executions.
To run the project set up desired config (Parallel or Sequential, number of threads, number of max iterations) in driver.py then run it.
The results will be save in the folder \results.
Module tsplib95 is a requirement for loading instances.
Instances were found at http://comopt.ifi.uni-heidelberg.de/software/TSPLIB95/
For initial solution it was used the Nearest Neighbour method.
For local search it was used the 2-opt method.
Sequential pertubation switches a random city and its right neighbour.
Parallel pertubation switches two random cities.
Some of the results found on instances berlin52 and st70.
