Synthetic Food Bank Network Dataset

A fully synthetic, openly licensed dataset describing a multi-stakeholder food bank network: clients with demographic attributes and modeled food preferences (stated and revealed), partner pantries with locations and operating hours, grocery-store and residential donors, volunteers with availability and delivery capacity, an inventory catalog, a stream of donation events, simulated pickup, delivery, and volunteer-shift histories, and driving-time origin-destination (OD) matrices connecting block groups, pantries, donors, and clients. The data is modeled on the West Alabama Food Bank nine-county service area but contains no real individuals, addresses, or identifying information. It is generated from open demographic inputs (American Community Survey tables and a fitted spatial point process) together with a preference model fit to a consented research survey. Only fitted model parameters, distributional summaries, and fully regenerated synthetic records are released, so the dataset can be inspected, regenerated, and reused without privacy constraints. Files. The dataset is provided as individual files (Apache Parquet tables plus JSON model-fit diagnostics) covering the population and network entities, service histories, spatial reference layers, stated and revealed preferences, fitted preference-model and travel-time-decay diagnostics, Huff utilization outputs, and origin-destination travel-time matrices, plus a single archive that contains the complete project (all data together with the code and documentation) for users who prefer one download. How to use the data. Runnable tutorials, per-file data dictionaries, and worked examples are maintained in the companion GitHub repository: https://github.com/UnivAL-IDA/synthetic-foodbank-network-dataset How to cite. If you use this dataset, please cite the accompanying Data Descriptor (Nature Scientific Data, in preparation; citation forthcoming). Until it is published, please cite this Zenodo dataset by its DOI. License. The dataset and its documentation are released under CC-BY-4.0; the accompanying source code (in the GitHub repository and the bundled archive) is released under the MIT License. Funded under the U.S. National Science Foundation SHARING program (Award No. 2125600). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.