Troubleshooting & FAQ
Troubleshooting
Attention
If you encounter problems, check these common issues first:
Solver not found → Make sure
glpsol(GLPK) orcplexis installed and available in your systemPATH.Adding to PATH:
On Linux/macOS: edit
~/.bashrc(or~/.zshrc) and add:export PATH=$PATH:/path/to/solverOn Windows: go to System Properties → Environment Variables → Path → Edit, then add the folder path containing
glpsol.exeorcplex.exe.Restart your terminal or IDE for changes to take effect.
Memory issues → Try solving with CPLEX (
--solver cplex) or increase available RAM.Long runtimes → Test the workflow with fewer years or a reduced temporal resolution before running the full scenario.
No LP/SOL file created → Double-check the input/output paths and confirm that the
results/directory exists.
Important
The AETOS model is computationally heavy. For a full-scale scenario such as AETOS_UNNZ20250911:
- Generating the LP file can take ~2 hours
- Solving with CPLEX on HPC can take ~6 hours
If your run fails or hangs, it is most likely due to insufficient memory, not an error in the workflow. Consider reducing the problem size or running on a machine with more RAM/cores.
If issues persist, please open an issue on GitHub: Report an Issue on GitHub
When reporting an issue, please include:
Step → At which stage of the workflow the error occurs (e.g. Step 1: Transform Input Data, Step 3: Solve with CPLEX)
Solver → Which solver was used (e.g. CPLEX) and version if known
System details → Operating system, Python version, and hardware specs (CPU/RAM, HPC vs local)
Logs → Full error message, warnings, or relevant log output (copy-paste or attach)
FAQ
Which solver should I use?
Use CPLEX if available (faster, more memory-efficient).
Where do I find the input data?
All processed inputs are hosted on Zenodo: AETOS Dataset on Zenodo
How do I install the model?
Follow the steps under Installation & Setup. Clone the repo, create an environment, and install either GLPK or CPLEX.
How do I run a quick test?
Run with smaller horizon outlook, or a smaller temporal resolution to ensure everything works before attempting the full model.
My model run is too slow, what can I do?
Try switching solvers, reduce the horizon outlook, or run on HPC with more memory.
Can I contribute improvements?
- Yes! Contributions are welcome. You can:
Open a pull request on GitHub
Submit an issue with your suggestion or bug report
Can I reuse AETOS data for my own research?
Yes, the data is open under a CC-BY license. Please cite the Zenodo repository and relevant publications.
How can I add a new country or technology?
You’ll need to extend the dataset and model definition. Check the documentation on Model Architecture before making changes.