Dataset and Code for Hybrid SARIMAX–PSO Inventory Optimization in Clinical Laboratories

his repository contains the complementary dataset and source code for the article “Hybrid SARIMAX–PSO Framework with Skew-Normal Residuals: Inventory Optimization in Clinical Laboratories”. The files included here allow full reproducibility of the experiments reported in the manuscript.

The repository is organized as follows:

• /data/

  • DemandaSpecialIssue.xlsx: original 36-month demand series for laboratory reagents.

  • PSO_ready_dataset.csv: processed dataset prepared for optimisation.

• /code/

•  

  • pso_inventory.py: implementation of the Particle Swarm Optimisation (PSO) algorithm with packaging and budget constraints.

• /results/

  • pso_solution.csv: optimal order quantities and associated costs for each product.

  • summary_metrics.csv: performance metrics (MAE, RMSE) comparing forecasting alternatives.

How to cite
If you use this dataset or code, please cite both this repository (Zenodo DOI) and the related article once published.

License
All code is released under the MIT License. The dataset is anonymized and provided for academic use only.