MATHEMATICAL OPTIMISATION prof. Lorenzo Castelli exam project by Massimo Trevi and Mike De Petris
Directory contents:
1-s2.0-S0377221722000078-main.pdf paper from https://www.sciencedirect.com/science/article/pii/S0377221722000078
Instances
directory with all input data files
Logs
directory with execution logs
2_input_from_files_as_paper.py model that reproduces paper results using the input data files given by paper authors
3_statistics_from_input_files.py processor program that parses and produces statistical data about the input data files, to understand frequency distributions and reproduce it to generate new scaled instances with random data
4_scale_random_instances.py model that finds optimization solutions for 20 random generated scaled instances
unlucky input.txt text file of instances selected from execution logs to investigate the model implementation behaviour with instances that give the worst optimal solutions
5_unlucky_files_create.py processor that creates input data files from "unlucky input.txt"
6_unlucky_case.py model that finds optimization solutions for the selected instances
7_scale_extended.py extended model that implements a new formulation with some hypothesis that keep in consideration the size of the goods to transport