Given cargos and trucks, find the optimal mapping to reduce distance, considering that is not possible to reuse trucks.
Matrix of geographical coordinates differences between of package pickup and truck location, combined with permutation of of cargo ordering based on index.
Suboptimal, considering that it's using all available permutations for range(len(cargos)) making the overall complexity O(len(cargos)!).
The solution was based on a brute-force implementation and could be improved if I had more time
Python 3.6.4
Linux Mint 18.3
$ python3.6 main.py
All the csv files are under data directory.
Each product has it's own section, with details about the selected truck and route and a generated link to google maps
In order to isolate environments, it's also possible to use Docker. Its already available a Dockerfile that could be used to run the solution:
$ docker build -t loadsmart-python . && docker run loadsmart-python
$ python3.6 -m unittest discover tests/ -p '*_test.py'