Lesson 2.4 — Forecasting Data Sources
Lesson 2.4 — Forecasting Data Sources
AI forecasting systems rely on high-quality data.
The quality of forecasting outputs depends heavily on the quality of inputs.
Common forecasting data sources include:
| Data Source | Example |
| Historical Sales | Previous customer demand |
| ERP Data | Orders and inventory |
| WMS Data | Warehouse movements |
| Transport Data | Delivery performance |
| Weather Data | Seasonal impacts |
| Market Data | Economic conditions |
| Supplier Data | Lead times and delays |
| Social Media Data | Demand sentiment |
| Web Analytics | Online customer activity |
Structured vs Unstructured Data
Structured Data
Structured data is organised and measurable.
Examples:
- SKU sales
- Inventory levels
- Lead times
- Forecast quantities
Unstructured Data
Unstructured data is more difficult to process.
Examples:
- Customer reviews
- Social media posts
- News reports
- Emails
Modern AI systems increasingly analyse both structured and unstructured data.