🌐 Language: English · Українська
End-to-end Power BI project analysing the warehouse inventory of a paint manufacturer & distributor: stock value, inventory turnover, slow-moving and dead stock, and an ABC classification of items. I built it to answer one question a warehouse or supply manager asks constantly — where is money tied up in stock, and which items deserve tight control?
Scope: within inventory (TMC), this project focuses on the goods-for-resale category. The same approach generalises to other inventory groups — raw materials, finished goods and spare parts.
🔗 The production-cost side of the same company is analysed in a separate project — Manufacturing Cost Analytics.
- What is the total stock value, and how does it change month to month?
- How fast does inventory turn over (inventory turnover, days of supply)?
- Which items are slow-moving or dead stock — where money sits frozen?
- How is stock value distributed by ABC class (where the value is concentrated)?
- How is stock split across warehouses and categories?
| Metric | Value |
|---|---|
| Annual consumption value | ≈ 16.7 M ₴ |
| Year-end stock value | ≈ 2.2 M ₴ |
| Inventory turnover | 12.7× / year |
| Days of supply | ≈ 29 days |
| Dead stock | 17.4 K ₴ across 4 SKUs |
| ABC split | A: 12 · B: 15 · C: 21 · D (dead): 4 |
The ABC analysis confirms a textbook Pareto: ~25% of SKUs drive ~80% of consumption value. Four items had zero movement all year — stock sitting on the shelf with money frozen — so I flagged them as a separate class D (dead stock).
- Power BI — data model (star schema), DAX measures, 3-page interactive report
- Power Query (M) — data cleaning, merges, and the full ABC classification pipeline
- Excel / CSV — source data
A star schema with two fact tables (Stock Movements, Inventory Snapshots) sharing three dimensions (Products, Warehouses, Calendar).
| Table | Role | Description |
|---|---|---|
Products |
Dimension | 48 goods with category, cost, price, supplier + ABC class |
Warehouses |
Dimension | 2 warehouses (Kyiv, Lviv) |
Calendar |
Dimension | 2024 date table (month, quarter, year) |
Stock Movements |
Fact | Receipts & issues transactions |
Inventory Snapshots |
Fact | Month-end stock on hand per item/warehouse |
The data transformation is the backbone of this project:
- Merged
unit_cost/categoryfromProductsinto the fact tables and computed value columns (movement_value,stock_value). - Built the ABC classification entirely in Power Query: filtered issues → grouped by SKU → sorted by consumption value → running total → cumulative % → conditional class (A ≤ 80%, B ≤ 95%, C otherwise).
- Broke a circular reference with an independent cost-lookup query — a real modelling problem, solved cleanly.
- Detected dead stock: items with no issues fall out of the ABC and were labelled class D, turning a
nullinto a business signal.
Total Issues Value · Closing Stock Value · Avg Stock Value · Inventory Turnover · Days of Supply · Dead Stock Value · Cumulative Consumption % (for the Pareto curve).
1. Overview — KPIs, stock-value trend, split by warehouse and category, key insights.
2. ABC Analysis — Pareto chart (bars + cumulative %), value by ABC class, top products.
3. Slow-moving & Dead Stock — dead-stock table (no movement), slow movers by days of supply, action recommendations.
├── data/
│ ├── products.csv
│ ├── warehouses.csv
│ ├── stock_movements.csv
│ └── inventory_snapshots.csv
├── inventory_warehouse.pbix # Power BI report (model + 3 pages)
├── inventory_warehouse_dashboard1.png
├── inventory_warehouse_dashboard2.png
├── inventory_warehouse_dashboard3.png
└── inventory_warhouse_data_model.png
Note on the data: figures are synthetically generated for demonstration and don't represent any real company. The assortment, seasonality and inventory behaviour, however, are modelled on how a real building-materials warehouse actually behaves.
Inna Tkachenko
If you found this project interesting or have feedback, feel free to reach out — I'm always open to a conversation.