Ant Colony Optimization applied to the Permutation Flow-Show Problem with Weighted Tardiness. The software provides implementation of the MMAS, M-MMAS and PACO algorithms, as well as support for bi-objective optimization.
The program can be called from the root of the project folder with a command of the form:
python run.py <path-to-instance> --method MMAS --time 30 --local-search swap --rho 0.4 --n-ants 40
where "method" can take the values "MMAS", "M-MMAS" and "PACO", "local-search" can take the values "swap", "interchange" or "insertion", rho is the pheromone trail persistence and n-ants is the number of ants in the colony.
For more command line arguments, simply type:
python run.py
- "run.py" is the entry point of the program.
- "hpo.py" is the script for hyper-optimizing the different algorithms.
NumPy, Numba
The project has been developped with the following versions:
- numpy == 1.15.4
- numba == 0.41.0