← Back to selected work
Hands-on reference implementation · Databricks
Spare Parts Inventory Optimization
A Databricks inventory-optimization implementation that converts spare-parts stock data into stockout-risk, reorder and excess-stock actions.
The problem
What this project is solving
Spare-parts teams need to protect availability without tying up unnecessary working capital. The project turns stock and lead-time data into explicit priorities instead of leaving planners with a generic inventory dashboard.
Implementation
What I built
- Created synthetic inventory data for 40 parts across two warehouses.
- Built Bronze, Silver and Gold Delta layers.
- Calculated lead-time demand, safety stock, reorder point, days of supply and inventory value.
- Classified items into High Stockout Risk, Reorder Soon, Excess Stock and Healthy, with recommended actions.
All quantities and values are based on synthetic reference data and are not presented as measured client savings.
Evidence
Reference results
6high stockout risk
4reorder soon
5excess stock
Delivery perspective
Why this matters
The implementation focuses on action: which items need attention first, what may need reordering and where excess inventory may be tying up value.