🔮 Forecasting Methods
1. Moving Average
Simple average of the last 3 periods. Best for stable demand patterns without trends.
2. Weighted Moving Average
Recent data weighted more heavily (50%, 30%, 20%). Better for recent changes.
3. Exponential Smoothing
Smooths out random fluctuations. Good for data with irregular patterns.
4. Linear Regression
Trend-based forecasting. Best when demand is growing or declining steadily.
5. Seasonal
Accounts for recurring patterns. Use when demand varies by month/season.
📊 Understanding Metrics
MAE (Mean Absolute Error)
Average prediction error. Lower is better.
RMSE (Root Mean Square Error)
Punishes large errors more. Lower is better.
MAPE (Mean Absolute Percentage Error)
Error as percentage. Less than 10% is excellent, 10-20% is good.
🆘 Troubleshooting
"Insufficient historical data"
Add more sales orders or wait for more demand history to accumulate.
"Unauthorized error"
Your session expired. Please log in again.
"Network error"
Check your internet connection and try again.