LAUSR

Abstract This paper tried to study the robot arm applied to the LCD panel manufacturing process. There is a fork structure in the robot arm used to perform the stretch into/back the glass-base board cassette. The box beam structure is used as the arm structure, and the carbon-fiber composite material is used to realize the feature of light weight. Multi-objective optimal design based on multi-performance characteristic index and response surface method is adopted to achieve two objective functions: static moving forward precision (SMFP) as well as dynamic moving forward and backward precision (DMFBP). Additionally, three control factors and two noise factors associated with the robot arm structure are selected to illustrate the first order statistical model with the RSM method. Case 1 studied the optimal static moving forward precision (SMFP); Case 2 studied the optimal dynamic moving forward and backward precision (DMFBP); Case 3 studied both SMFP and DMFBP simultaneously. The findings indicate that the optimal belt tension values are obtained for the robot arm to perform the optimal precision under static and dynamic operation; what's more, it is verified that the optimal statistical model can provide a good reference for creating a cloud model to improve the manufacturing process in the era of industry 4.0.