Instances and Code of the Paper "The Path-Precedence Constrained Knapsack Problem with Unit Weights: Exact Algorithms, Linear Formulations and Instances"
Description
These files have all the data and code required to reproduce the experiments presented in the paper "The Path-Precedence Constrained Knapsack Problem with Unit Weights: Exact Algorithms, Linear Formulations and Instances". That scientific paper is available for free in the proceedings of the LVI Brazilian Symposium on Operations Research (SBPO 2024) and may be cited as: Monteiro, Cristiano Martins; Figueiredo, Tatiane Fernandes; del Priore, Joaquín; Xavier, Eduardo Candido; Negrotto, Daniel. The Path-Precedence Constrained Knapsack Problem with Unit Weights: Exact Algorithms, Linear Formulations and Instances. In: Anais do LVI Simpósio Brasileiro de Pesquisa Operacional; 2024. Available at: https://proceedings.science/sbpo/sbpo-2024/papers/the-path-precedence-constrained-knapsack-problem-with-unit-weights-exact-algorit?lang=en. This dataset is composed by two files: - The file PathPrecedenceConstrainedKnapsack-UnitWeights.ipynb is a Jupyter notebook with all the algorithms and formulations proposed in the paper. - The file instances_ppckp-uw.xlsx has all the instances described in the paper. The first worksheet presents the name of the instances, the capacity value considered and the optimal value obtained. The other worksheets show the data for each instance. The source codes, the instances and the scientific paper were made by the same authors. All authors work at the Shipping Optimization Team (SOT), which is part of the Shipping Operational Research Team (SOR) from Mercado Libre (Argentina) and Mercado Livre (Brazil).
Files
Steps to reproduce
In order to run the Jupyter notebook, it is required to have the a Python3 installed and have the following dependencies: Jupyter, openpyxl (for reading the excel file) PuLP 2.8.0 (for running the linear formulations).