Allocation of Flexible Tooling for Optimal Stochastic Multistation Manufacturing Process Quality Control

Stream of Variance (SoV) modeling of multi-station manufacturing process has been studied for the past 15 years and was used for identification of root causes of manufacturing errors, characterization and optimal allocation of measurements in multi-station manufacturing processes, process-oriented tolerance allocation, and most recently, for optimal in-process adaptations of programmable, flexible tooling (flexible fixtures, CNC machines) for autonomous minimization of errors in dimensional product quality. However, due to the high cost of flexible tooling, it is plausible to strategically position such devices across a manufacturing system in a way that one’s ability to mitigate quality problems is maximized. In this paper, a distributed stochastic feed-forward control method is devised which gives the optimal (in least square sense) reduction of the variance-covariance matrix of errors in dimensional workpiece quality in a multi-station manufacturing process with a limited number of flexible tooling components. Genetic Algorithm is proposed to enable optimal allocation of flexible tooling devices. Theoretical results have been evaluated and demonstrated using the SoV model of a real industrial process.