A Ayam Merak 's coffee powder 75 gram is the product that is focused for this research because it causes major loss problem for the company. The inspection standard is observed based on the attribute aspects and the variable aspects. From the calculation result, the non­ conforming proportion from the attribute aspects is normal for the company is at 9.3%. In addition from the variable aspect, some of the aspects already have a fine capability process such as weight of the product after production, the water content and caffeine content before fried process . However some of the variable aspect still have some low capability process such as the weight of raw coffee bean (Cp = 0.85, Cpk = 0.88, Cpm = 0.33), weight of friedcoffee bean (Cp = 0.44, Cpk = 0. 19, Cpm = 0.35), caffeine content after fried (Cp = 0.51, Cpk= 0.37, Cpm = 0.47), dan water content after fried (Cp = 0.71, Cpk = 0.21, Cpm = 0.39). Based on the above criteria and consideration, the research is narrowed on the low-capabilityvariable aspects. From research and interview session, it is known that the deviation occurs because the inappropriate machinery setting process is only based using try and error method . Therefore, Multi Response Surface Method is used for obtaining the machinery setting for optimal performance .The Analysis is performed on the uncooked material storage process at the warehouse . Based on analysis data in the screening factor phase using full factorial design method be known there are two critical factor that significantly influencing the weight ofraw coffee bean that is Factor A (Temperature) and Factor C (humidity). Since these two factors are far away from the optimum region, the Steepest Ascent method is used. After the experiment region is on the optimum area and a curvature is identified, Central Composite Design is used . The optimum machine settings that can be proposed are Temperature at 27.4 °C, air velocity at 1.2 m/sec, and humidity at 22%. Furthermore experiment is continued with testing on other factor on weight, caffeine content and water content of raw coffee bean with Evolutionary Operation method . The result of the method shows that other factors only have minimal impact on the three response variable factors. The new optimum setting then is implemented and obtainedcapability process for weight (Cp = 1.29, Cpk = 1.16, Cpm = 1.21), caffeine contP,nt of rawcoffee bean (Cp = 1.5, Cpk = 1.32, Cpm = 1.32), and water content ofraw coffee bean (Cp =1.36, Cpk = 0.91, Cpm = 0.81).Additionally the research is continued on the frying process with Screening Factors method and obtained factors that influencing the weight, water content, and caffeine content after the coffee is fried that is Factor B (Temperature) and Factor C (Process Time). Before using the Multiple Response Surface method to set the factor, Central Composite method is implemented to identify the optimum region . The optimum setting for the machinery are Factor A (rotation speed) at 15 rotation I minute, Factor B (Temperature) at 232 °C and Factor C (Process Time) at 37.5 minutes. The result of the implementation based on optimum setting on the fried machinery has given the high capability process on the three response variablethat is weight of fried coffee bean (Cp = 1, Cpk = 0.74, Cpm = 0.78), the water content offried coffee bean (Cp = 1.04, Cpk = 0.88, Cpm = 0.94), the caffeine content of fried coffeebean (Cp = 1.51, Cpk = 1.25, Cpm = 1.19)