A critical aspect for the long-term success of research in computational optimisation is the availability of libraries of benchmark instances. In MINOA, we generate and collect such instances that are difficult enough to challenge the state of the art and can drive new solution methods.
The instances are of varying difficulty for the various MINO applications. The benchmarks are made publicly available, mainly through the existing platform MINLPLib, in order to reach large visibility.
The goal of this collection is to drive further improvements in algorithmic research by providing the chance to compare new solution methods to the ones developed within MINOA. Also, those instances reflect the difficulties and the characteristics of the underlying real-world tasks so that practitioners can learn about the solvability of the problems.
This library is updated continuously throughout the project duration.
Link to MINLPLib: http://www.minlplib.org/
(A list of the instances the MINOA researchers contributed will be available soon.)
We also provide instance generators or large collections of instances for specific applications:
-
-
- Pierre Bonami, Andrea Lodi, and Giulia Zarpellon (2018): Learning a classification of mixed-integer quadratic programming problems
→ Download experimental data and manual
→ Download article - Lotka Volterra fishing problem
→ Download experimental data and manual - M. Jünger, E. Lobe, P. Mutzel, G. Reinelt, F. Rendl, G. Rinaldi, T. Stollenwerk (2019): Performance of a Quantum Annealer for Ising Ground State Computations on Chimera Graphs
→ Download article (pdf)
→ Download experimental data
→ Structure of the data
- Tran Bao Duy (2019): Time optimal car problem
→ Download experimental data and manual - Tran Bao Duy (2019): Van der Pol Oscillator (binary variant)
→ Download experimental data and manual - Martina Cerulli (2021): Aircraft deconfliction problem via subliminal speed regulation in 3 dimensions
→ Download experimental data and manual - Shudian Zhao (2021): K-equipartition problems and graph partition problems with knapsack
constraints
→ Download experimental data and manual
- Pierre Bonami, Andrea Lodi, and Giulia Zarpellon (2018): Learning a classification of mixed-integer quadratic programming problems
-