Response Surface Method Approach for Optimizing Roasting Condition in Robusta Coffee Cupping Test Quality

Abstract

The Roasting process was an important step to generate good-quality roasted coffee beans. Two factors that affected coffee bean perfection were roasting time and temperature. An appropriate time and temperature will emphasize the greatest coffee bean aroma, color, and flavor, while the speed of the roasting process depends on the number of stirring fins in a roasted machine. The purpose of this study was to optimization time, temperature, and stirring fin number in the roasting process to generate the best cupping test roasted coffee beans. The Central Composite Design (CCD) was used in this study to determine Response Surface model. Minitab 17 and Design Expert 4.0 software used to determine the combination for all variables. Robusta green bean was taken from Kota Agung Timur, Lampung, which had density of 701—750 g/L. The result recommended roasted coffee bean was at 213oC for 16.7 minutes and used 3 stirring fins. This condition will generate cupping test score 84.92 which is define as Excellent in score notation criteria. CCD methods has develop an equation from optimal process to predict roasted coffee bean cupping test quality i.e., Y = 85.6 – 0.0088X1 – 1.2X2 – 11.06X3 + 0.000043X12 – 0.038X22 – 0.70X32 + 0.01000X1X2 + 0.0625X1X3 + 0.300X2X3.

Keywords

• cupping test
• response surface model optimization
• robusta coffee bean.