Computational Logistics of the Vehicle Routing Problem with Time Windows (VRPTW). Comparative Review of application of the solution techniques, the Particle Swarm Optimisation (PSO) algorithm and Genetic Algorithm (GA) to the VRPTW.
Conference Proceedings: https://ieeexplore.ieee.org/document/9004294
The solution technique algorithms are based according to the respective references given below. The algorithms are coded using Python 3.
Ombuki, Beatrice, Brian J. Ross, and Franklin Hanshar. "Multi-objective genetic algorithms for vehicle routing problem with time windows." Applied Intelligence 24.1 (2006): 17-30. https://link.springer.com/content/pdf/10.1007/s10489-006-6926-z.pdf
Gong YJ, Zhang J, Liu O, Huang RZ, Chung HS, Shi YH. Optimizing the vehicle routing problem with time windows: a discrete particle swarm optimization approach. IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews). 2012 Mar;42(2):254-67. https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=5773510
https://www.sintef.no/projectweb/top/vrptw/solomon-benchmark/
For each of the algorithms and the respective metrics ('../VRPTW_code/algorithm_Metric'), the code for the initial encoding and optimisation technique can be found in the following files:
Initial Encoding: '../initial_encoding/main_initialEncoding.py'
Optimisation: '../main/main.py'
Initial Encoding: '../initial_encoding/initialEncoding_case1.py'
Optimisation: '../main/main.py'
NOTE: Initial encoding needs to be run before running the main file in order to create the initial candidate solutions.
Please note that the 'initial_encoding/initial' folder contains the initial encodings, and the '/main/final' folder contains some experiments from solving the VRPTW for datasets with 25 customers. These folders would need to be replicated to run the datasets with 50 and 100 customers. And set the 'num_customers' in the respective files to the dataset that you are considering i.e. 25, 50 and 100 customers.