DVRP-FPTW-instances

In this repostory there are the benchmark instances to test the DVRPFPTW.

Dependecies

Works with Python 3.8+ and only depends on numpy.

Example usage

To read the instances you must import numpy libary for python and following the next procedure:

import numpy as np

# Read Static benchmark instance and solution
static_instance = np.load(instance_file, allow_pickle='TRUE').item()
static_solution = np.load(instance_file_sol, allow_pickle='TRUE').item()

instance and solution are dictionaries that contain all parsed data.

>>> static_instance.keys()
dict_keys(['name', 'n_customers', 'node_coord', 'demand', 'service_time', 'edge_weight', 'revenue', 'vehicles', 'capacity', 'autonomy', 'time_window'])

>>> static_solution.keys()
dict_keys(['Name', 'Routes', 'Min_profit', 'Total_profit'])

# Read Dynamic benchmark instance and solution 
dynamic_instance = np.load(dynamic_instance_file, allow_pickle='TRUE').item()
dynamic_solution = np.load(dynamic_instance_file_sol, allow_pickle='TRUE').item()

>>> dynamic_instance.keys()
dict_keys(['name', 'static_instance', 'static_solution', 'event'])

>>> dynamic_instance["event"] 
{'type': 'Vehicle_breakdown',
 'broken_vehicle': 'vehicle_12',
 'broken_time': 28.178005607210743}