All projects Animated overview of the steps below (an illustration, not recorded footage).
Steel demand forecasting & inventory planning
Monthly HRC and structural steel forecasts that guide stock decisions
- Time-series forecasting
- Microsoft Excel
- Python
Before vs after
Before
Before
- Bottlenecks in procurement and inventory workflows
- No clear view of monthly demand
After
After
- Monthly demand forecasts per product
- Reorder point, safety stock and DIO
From sales history to stock decisions
01Step 1 of 4
Map workflows
Mapped procurement and inventory workflows to find bottlenecks.
02Step 2 of 4
Sales history
Monthly demand history for hot-rolled coil and structural steel.
03Step 3 of 4, Time-series forecasting
Forecast models
Moving Average, Exponential Smoothing and ARIMA predict monthly demand.
04Step 4 of 4
Inventory policy
Reorder points, safety stock and DIO support stock decisions.
Tech stack in this project
- Time-series forecastingMoving Average, Exponential Smoothing, ARIMA
Why it matters for supply chain
- Forecasting = reorder points and safety stock
- DIO = working capital tied up in stock
- Workflow mapping = procurement bottlenecks found early
Linked to my Master of Logistics & SCM, University of Sydney
Read the full case study
Problem
Procurement and inventory decisions for steel products needed a clearer view of expected monthly demand and of where bottlenecks sat in the procurement and inventory workflow.
Approach
- Mapped procurement and inventory workflows to identify bottlenecks.
- Developed time-series demand forecasting models (Moving Average, Exponential Smoothing and ARIMA) to predict monthly demand for hot-rolled coil and structural steel products.
- Applied inventory planning models (reorder points, safety stock, days inventory outstanding) to support stock decisions.