David Bauer

Inventory optimization software

I built software that decides what to stock, how much, and when to reorder. It prioritizes the product portfolio, sets a service level per SKU, and balances demand against supply — so the business avoids stockouts on the products that matter without tying up capital in dead stock.

What it does

• Portfolio prioritization: Ranks SKUs by sales performance and strategic importance, so money and attention go to the products that actually pay.
• Service levels: Sets a target probability of not running out during a replenishment cycle, per SKU — high for the A-items, lower for the long tail.
• Demand forecasting: Forecasts per-SKU demand with fbProphet and flags suspicious data before it poisons the forecast.
• Stock level calculation: Computes safety stock, reorder point, and maximum stock from SKU importance, supplier reliability, lead time, and forecast error.
• Supplier analysis: Tracks each supplier's lead times, reliability, and turnover — including keeping order volumes above supplier-bonus thresholds.
• Slow movers: For low-demand items, suggests bundling with popular products or finding another market instead of just reordering less.
• Order timing and quantity: Tells you when to order and how much, which is the whole point.

The outcome: fewer stockouts, less capital sitting in the warehouse, and orders placed at the right time in the right quantity. Built in .NET with Apache Kafka for the data flow and Python (fbProphet) for the forecasting.