Two-dimensional pheromone propagation controller applied to run-to-run control for semiconductor manufacturing

Abstract

This paper presents a new perspective on process control, called the two-dimensional pheromone propagation controller (2D-PPC), which considers the spatial information about disturbances of the process within a wafer to generate new predicted intercepts of the models for the subsequent use in time-effect controller (the exponentially weighted moving average, EWMA, in this study). The 2D-PPC assumes that the disturbances have their own behavior and affect others nearby in a wafer at a run; thus, it involves the "space-effect" among disturbances of the process at measurement positions within a wafer. The framework of the space-time controller (STC), which interlaces the time-effect controller and the space-effect 2D-PPC is constructed, and the stability analysis and intrinsic characteristics of the STC are discussed. Simulations are conducted using two-dimensional anthropogenic disturbances generated from fabrication data. The results show that the STC has better performance as compared to the conventional time-effect controllers. From implementation view point, since STC does not change the original code of time-effect controller, it can be easily implemented in the current process control loop by only adding an additional space-effect controller.