📘 How To Use AI Inventory Demand Forecaster
- 📊 Enter Sales Data: Provide historical sales in JSON or CSV format (include months, sales, optionally returns).
- ⚙️ Select Model: Choose between Simple Moving Average, Trend-Adjusted, or Seasonal (basic) forecast.
- ⏱️ Set Parameters: Input lead time (days) and desired service level (%) for safety stock calculation.
- 🔮 Generate Forecast: Click "Generate Forecast" to compute reorder points, stockout risk, and overstock reduction potential.
- 📋 Review & Copy: Analyze key metrics and copy results for your planning.
💡 Pro Tips:
- Use at least 6-12 months of data for better accuracy.
- Higher service level (95%+) increases safety stock but reduces stockout risk.
- The tool estimates overstock reduction by comparing current stock (assumed) to optimized reorder levels.
- Results are client-side; your data stays private.
🔍 Example Input (JSON)
[
{"month": "Jan", "sales": 120, "returns": 5},
{"month": "Feb", "sales": 135, "returns": 7},
{"month": "Mar", "sales": 110, "returns": 4}
]
❓ Frequently Asked Questions
How is the reorder point calculated? ▼
We use: Reorder Point = (Average Daily Sales × Lead Time) + Safety Stock. Safety stock is based on demand variability and your service level.
What does "stockout risk" mean? ▼
It's the probability that demand will exceed available stock during lead time, given current inventory levels. Lower service levels increase this risk.
How accurate is the overstock reduction estimate? ▼
The 15-20% reduction is an industry benchmark. Your actual savings depend on data quality, lead time consistency, and demand patterns. The tool gives a projection based on your inputs.
Can I use CSV data? ▼
Yes! Select the CSV option and paste data in format: month,sales,returns (returns optional). Headers required.
Is my sales data secure? ▼
Absolutely. All calculations are performed in your browser. We never send or store your data on any server.
What models are used? ▼
Simple Moving Average, basic trend-adjusted, and a naive seasonal model. These are lightweight and work well for many small to medium inventories.