Grain loss due to combine header comprises more than 50% of the entire harvesting losses. This study was performed to estimate and predict the combine header loss at different adjustments of combine header components using regression analysis and neuro-fuzzy model. Reel index (forward speed of combine harvester devided by peripheral speed of reel), cutting height, the horizontal distance of reel tine bar from cutter bar and vertical distance of reel tine bar from cutter bar were considered as input variables and combine header loss was regarded as output variable. Neuro-fuzzy model showed the coefficient of determination (R²) equal to 0.9817 which is superior to multiple regression method with 0.6354.
