Anonymised lot-sizing and storage-footprint dataset with reproducibility code for injection molding

This record provides the anonymised data and reproducibility code accompanying the manuscript “Lot sizing with storage-footprint cost in injection moulding: a simulation-based DSS, its decisions, robustness and limits.” The industrial dataset contains 17 pseudonymised production run units. It reports normalised storage footprints, machine classes, setup-cost indices, annual demand, evaluated lot-size decisions and the corresponding run counts. Company identifiers, product and part identities, order numbers and absolute costs have been removed. The accompanying Python script reproduces the analytical lot-sizing comparisons, robustness experiments, inventory-trajectory analysis, peak storage-occupancy results and dispersion criterion reported in the manuscript. It also generates the included results for 2,640 synthetic portfolios using the documented random seed. Files: supplementary_dataset.csv — anonymised and normalised industrial input data and evaluated lot-size decisions supplementary_analysis.py — Python reproducibility script synthetic_dispersion_study.csv — generated results for 2,640 synthetic portfolios README.md — file descriptions, requirements and execution instructions The analysis requires Python with NumPy and pandas. Random seeds, modelling conventions and the interpretation of the reported indices are documented in the analysis script. Absolute industrial costs, original ERP records, product identities and the proprietary Siemens Plant Simulation model are not included because they are confidential. Licensing: the CSV files and documentation are licensed under CC BY 4.0. The Python analysis script is licensed under the MIT License.