Loading...
Loading...
Contributors:
Dataset URI:
w6zjtdjk9mThis dataset contains transaction-level sales records and stock movement records from a retail pharmacy in Central Java, Indonesia, covering 2 January 2023 – 31 December 2025 (no records exist on a small number of days, e.g. New Year's Day and Eid al-Fitr; see README). The records were exported from the pharmacy's point-of-sale (POS) system and are provided in two forms: raw exports and cleaned, analysis-ready versions. Column headers are translated from Indonesian to English; Indonesian terms such as Penjualan (sale), Pembelian (purchase) and Mutasi (stock transfer) remain in the data values. FILES: - raw_dataset/sales_raw.csv: 649,656 sales rows, 35 columns. Values as exported, with personal names anonymized. - raw_dataset/stock_raw.csv: 804,027 stock movement rows, 15 columns. Values as exported. - processed_dataset/sales_processed.csv: 627,389 sales rows, 35 columns. - processed_dataset/stock_processed.csv: 775,560 stock movement rows, 21 columns. - processed_dataset/EDA Notebook/EDA_Analysis.ipynb: exploratory data analysis notebook and figures. - processing script/generate_raw.py and generate_processed.py: Python scripts that produce the raw and processed CSVs. - README.md: column dictionary and file guide. PROCESSING APPLIED (raw to processed): - Date and time split into Date, Year, Month, Day, DayOfWeek, Time and Hour. - Numeric fields converted to numeric types (e.g., percentages such as "8%" become 8.0; stock quantities such as "17 Fls" become the integer 17 with the unit in a separate Unit column). - Combined batch/expiry text split into BatchNumber and ExpiryDate. - Free-text stock transaction descriptions (140+ spelling variants) normalized into TransactionType. Purchases (Pembelian) keep the supplier name in brackets. - Categories excluded from the processed files: Alkes (medical/health equipment), Baby, and Minuman (beverages). They remain in the raw files. - Seven sales columns dropped due to near-zero fill rates or privacy concerns: PrescriptionNumber, PatientID, DoctorID, SalespersonID, SalesPromo, PromoName, TaxInvoiceNumber. - The Warehouse column dropped from stock (a single constant value). PRIVACY: Cashier, patient, doctor and salesperson names were replaced with anonymized codes (e.g., Cashier_001) before publication. The mapping to real names is not included. The dataset contains no cost or purchase-price data: PriceTier is a categorical code for the selling-price tier. POSSIBLE REUSE: Demand forecasting, inventory and expiry-risk analysis, sales seasonality studies, drug-class and product-mix analysis, and data-cleaning or machine-learning benchmarking on real-world Indonesian retail pharmacy data.
Keyword(s):
Pharmacy; Time Series.
MeSH Term(s):
Grants and Funding:
No Grants or Funding Data Available
Associated Publication(s):
No Associated Publication Data Available
Supplementary Data:
No Supplementary Data Available
Related Dataset(s):
No Related Dataset Data Available
The National Library of Medicine (NLM) provides information about biomedical datasets. Inclusion on an NLM Web page does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: NLM Web Policies (nih.gov)