Wrong max fitness value

iRB-Lab / py-ga-VRPTW

A Python Implementation of a Genetic Algorithm-based Solution to Vehicle Routing Problem with Time Windows

MIT License

560 stars 170 forks source link

Wrong max fitness value #11

Closed radoslawik closed 4 years ago

radoslawik commented 4 years ago

Hey, I have been trying the code and I discovered that max_fitness is not changing within number of generations. It makes the result always the same, even if there is only 1 generation. Firstly I was convinced that it can be library versioning or something connected to that, but the issue is also visible in the result csv's (max_fitness column does't seem to change).

iROCKBUNNY commented 4 years ago

Hey @radoslawik ! Thanks a lot. Can you fix it?

radoslawik commented 4 years ago

I have just noticed that my objection is the copy of #5 Performance Issue. The result is a random route that always seems to be is in initial population, even if population size is really small (for example 20).

iROCKBUNNY commented 4 years ago

@radoslawik The algorithm is not tuned and the performance is definitely poor. This was an experimental project. Please take it as a very fundamental (and far from robust) framework for your research work. With regards!