MULTIVARIATE ECONOMIC-ANALYSIS OF ROBOT PERFORMANCE REPEATABILITY ANDACCURACY

This paper presents a multivariate linear statistical model used in a three-factor experiment to investigate the effects of speed, traveling distance, and part weight on robot repeatability and accuracy. An emp irical regression fit obtained showed that repeatability was influence d only by manipulator speed, while accuracy was influenced by manipula tor speed and part weight for the test robot and levels of experimenta l factors considered. Furthermore, statistical analysis produced an ec onomic decision for optimizing the long-run expected profit per unit o f time. The empirical study is based on the simulation of a multistati on assembly operation with a PUMA robot.