“Ingredients based forecasting methodology”, which was proposed by Doswell et al. in 1996, is briefly presented. By comparing with traditional “pattern recognition” method, the strength and weakness of the “ingredients based forecasting methodology” are described. With real thunderstorm initiation case, the application of “ingredients based forecasting methodology” is illustrated. It is emphasized that “ingredients based forecasting methodology” and traditional “pattern recognition” forecasting method complement each other. Some misunderstandings on the “ingredients based forecasting methodology” are clarified.