Example: Aging Workflow (Storage Stability)¶
Goal: Model degradation trajectories over storage time (days/months) and estimate remaining shelf-life.
🔬 Minimal Reproducible Example (Synthetic)¶
from examples.aging_quickstart import main
# Runs synthetic dataset generation, trajectory fitting, and shelf-life estimation
main()
Expected Output:
Dataset: ~300 spectra, time range 0-180 days
Entities: 5
Trajectory metrics (first 5 rows):
entity slope acceleration
... ... ...
Saved: outputs/aging_degradation_trajectories.png
Shelf-life estimates (days):
entity t_star ci_low ci_high slope intercept
... ... ... ... ... ...
Saved: outputs/aging_shelf_life_estimates.png
📈 Figures¶
- Degradation Trajectories: docs/assets/workflows/aging/degradation_trajectories.png
- Shelf-Life Estimates: docs/assets/workflows/aging/shelf_life_estimates.png
- Residual Diagnostics: docs/assets/workflows/aging/residual_plot.png