Present_Wrapping_Problem

CP, SAT and SMT implementation of the PWP.

Authors

• Lorenzo Mario Amorosa
• Mattia Orlandi
• Giacomo Pinardi

Project structure

The project is structured in 5 folders:

• CP, which contains the report and the code related to the CP implementation;
• SAT, which contains the report and the code related to the SAT implementation;
• SMT, which contains the report and the code related to the SMT implementation;
• Instances, which contains all the 33 instances in .txt files;
• pwp_utilities, which contains three python scripts:
  • instance_to_dzn.py to convert the instance from .txt to .dzn (usage: python instance_to_dzn.py <input_file> <output_file>);
  • plot_solution.py to generate an image of a given solution (usage: python plot_solution.py -f <file_name>);
  • solve_all_cp_instances.py to generate one solution for each instance (to know how to use it: python solve_all_cp_instances.py --help).

CP

The CP folder contains 4 subfolders:

• benchmark, containing all the bash scripts to run the benchmarks and all the corresponding results in .txt (you must run them from the CP folder);
• in, containing the instances in .dzn;
• out, containing the solutions to the instances in .txt;
• src, containing all the models developed:
  • the main ones:
    • the final one is pwp-final.mzn;
    • the one for taking into account rotation is pwp-rotation.mzn;
    • the one for taking into account identical pieces is pwp-identical.mzn;
  • the naive ones:
    • the naive model is pwp-naive.mzn;
    • the dual model is pwp-dual.mzn.

SAT

The SAT folder contains 3 subfolders:

• benchmark, containing all the bash scripts to run the benchmarks and all the corresponding results in .txt (you must run them from the SAT folder);
• out, containing the solutions to the instances in .txt;
• src, containing all the models developed:
  • the main ones:
    • the final one is pwp-final.py (to know how to use it: python pwp-final.py --help);
    • the one for taking into account rotation is pwp-rotation.py (to know how to use it: python pwp-rotation.py --help);
    • the one for taking into account identical pieces is pwp-identical.py (to know how to use it: python pwp-identical.py --help);
  • the corresponding Jupyter notebooks used for data exploration:
    • the final one is pwp-final.ipynb;
    • the one for taking into account rotation is pwp-rotation.ipynb;
    • the one for taking into account identical pieces is pwp-identical.ipynb.

SMT

The SMT folder contains 3 subfolders:

• benchmark, containing all the bash scripts to run the benchmarks and all the corresponding results in .txt (you must run them from the SMT folder);
• out, containing the solutions to the instances in .txt;
• src, containing all the models developed:
  • the main ones:
    • the final one is pwp-final.py (to know how to use it: python pwp-final.py --help);
    • the one for taking into account rotation is pwp-rotation.py (to know how to use it: python pwp-rotation.py --help);
    • the one for taking into account identical pieces is pwp-identical.py (to know how to use it: python pwp-identical.py --help);
  • the corresponding Jupyter notebooks used for data exploration:
    • the final one is pwp-final.ipynb;
    • the one for taking into account rotation is pwp-rotation.ipynb;
    • the one for taking into account identical pieces is pwp-identical.ipynb.