Object-oriented design and implementation of an optimizing run-to-run quality controller for semiconductor manufacturing processes.

摘要

This dissertation presents a new run-to-run (R2R) multiple input-multiple output (MIMO) controller for semiconductor manufacturing processes. The controller, termed Optimizing Adaptive Quality Controller (OAQC), can act both as an optimizer--in case equipment models are not available--or as a controller for given models. Recursive estimation techniques are utilized for the on-line estimation of the parameters of a MIMO model of the Hammerstein type. This model allows control of linear and nonlinear systems.;An object-oriented design (OOD) is developed in this dissertation for the software implementation of the new R2R controller. The OOD is generic and can be utilized for the integration of any MIMO Engineering Process Control (EPC) scheme with any multivariate Statistical Process Control (SPC) chart. The SPC chart can be used to detect deviations from target or act as a deadband on the EPC scheme. By applying this OOD method, the software design of the OAQC is described.;The software implementation of the OAQC using the NeXTSTEP object-oriented operating system is described. The NeXTSTEP OAQC can run in "standalone" form or integrated with the University of Michigan's Generic Cell Control (GCC), a discrete-event controller for use in semiconductor manufacturing processes.;For testing purposes, four Chemical Mechanical Planarization (CMP) processes were simulated based on real equipment models provided by SEMATECH and other semiconductor companies. Both the open loop and closed loop behaviors of the proposed R2R controller were studied. It is shown that under the presence of noise and drift disturbances of different magnitudes, the OAQC allows to keep adequate control of the responses even if the input-output transfer function is severely nonlinear.