online-spp

Implements Thompson Sampling Algorithm using a Gaussian Field distribution assumption over arcs velocity From the paper: "The Online Shortest Path Problem: Learning Travel Times Using a Multiarmed Bandit Framework" https://doi.org/10.1287/trsc.2023.0196

Execute the codes inside the project folder.

Run real life instance to replicate the results in the paper:

Windows:

FOR /L %i IN (0,1,99) DO python main_from_paths.py 150 %i

Linux:

for number in {0..99}; do python main_from_paths.py 150 $number; done

Data:

https://github.com/felipelagos/beijing-instance.git

Folder containing the codes should look like:

/path_to_folder/online-spp -> project folder

/path_to_folder/data/new_large/nodes.csv: format -> id;latitude;longitude

/path_to_folder/data/new_large/arcs.csv: format -> id;node1;node2;times

Run artificial instances to replicate the results in the paper:

Windows:

FOR /L %i IN (1,1,4) DO FOR /L %j IN (1 3 4) DO python main.py 50 51 %i 0 %j 500
FOR /L %i IN (1,1,4) DO FOR /L %j IN (1 3 4) DO python main.py 150 151 %i 0 %j 20

Linux:

for number1 in {1..4}; do for number2 in {1,3,4}; do python main.py 50 51 $number1 0 $number2 500; done; done
for number1 in {1..4}; do for number2 in {1,3,4}; do python main.py 150 151 $number1 0 $number2 20; done; done