Cost function so we can compare algorithms:

def calculate_total_kwh_m(mst_table): return (mst_table['routes']*abs(mst_table['heat_total'])).sum()

Code: https://kdrive.infomaniak.com/app/share/745832/90ca4574-d096-4672-bf44-db07a58fb955/files/3934

Iterative optimization with random perturbation of one distance.Reduction in total cost sum(abs(heat_total)*routes):

requirements.txt content extracted by github copilot:

numpy pandas geopandas pickle-mixin sparse xarray scipy matplotlib contextily