Thesis (Ph. D.)--University of Florida, 2001. / Title from first page of PDF file. Document formatted into pages; contains viii, 118 p.; also contains graphics. Vita. Includes bibliographical references (p. 115-117).

Thesis (Ph. D.)--Rutgers University, 2009. / "Graduate Program in Industrial and Systems Engineering." Includes bibliographical references (p. 114-120).

Thesis (Ph. D.)--University of Texas at Austin, 2002. / Vita. Includes bibliographical references. Available also from UMI Company.

Thesis (Ph. D.)--University of Texas at Austin, 1999. / Vita. Includes bibliographical references (leaves 102-106). Available also in a digital version from Dissertation Abstracts.

Includes bibliographical references (p. 64-71).

Thesis (Ph. D.)--University of Hong Kong, 2002. / Includes bibliographical references (leaves 196-209).

Thesis (Ph. D.)--Drexel University, 2002. / Includes abstract and vita. Includes bibliographical references (leaves 104-109).

Thesis (Ph. D.)--Lehigh University, 2003. / Includes vita. Includes bibliographical references (leaves 133-143).

This thesis consists of two topics: statistical selection and profile monitoring. Statistical selection is related to ranking and selection in simulation and profile monitoring is related to statistical process control. Ranking and selection (R&S) is to select a system with the largest or smallest performance measure among a finite number of simulated alternatives with some guarantee about correctness. Fully sequential procedures have been shown to be efficient, but their actual probabilities of correct selection tend to be higher than the nominal level, implying that they consume unnecessary observations. In the first part, we study three conservativeness sources in fully sequential indifference-zone (IZ) procedures and use experiments to quantify the impact of each source in terms of the number of observations, followed by an asymptotic analysis on the impact of the critical one. Then we propose new asymptotically valid procedures that lessen the critical conservativeness source, by mean update with or without variance update. Experimental results showed that new procedures achieved meaningful improvement on the efficiency. The second part is developing a wavelet-based distribution-free tabular CUSUM chart based on adaptive thresholding. WDFTCa is designed for rapidly detecting shifts in the mean of a high-dimensional profile whose noise components have a continuous nonsingular multivariate distribution. First computing a discrete wavelet transform of the noise vectors for randomly sampled Phase I (in-control) profiles, WDFTCa uses a matrix-regularization method to estimate the covariance matrix of the wavelet-transformed noise vectors; then those vectors are aggregated (batched) so that the nonoverlapping batch means of the wavelet-transformed noise vectors have manageable covariances. Lower and upper in-control thresholds are computed for the resulting batch means of the wavelet-transformed noise vectors using the associated marginal Cornish-Fisher expansions that have been suitably adjusted for between-component correlations. From the thresholded batch means of the wavelet-transformed noise vectors, Hotelling’s T^2-type statistics are computed to set the parameters of a CUSUM procedure. To monitor shifts in the mean profile during Phase II (regular) operation, WDFTCa computes a similar Hotelling’s T^2-type statistic from successive thresholded batch means of the wavelet-transformed noise vectors using the in-control thresholds; then WDFTCa applies the CUSUM procedure to the resulting T^2-type statistics. Experimentation with several normal and nonnormal test processes revealed that WDFTCa outperformed existing nonadaptive profile-monitoring schemes.

Monitoring and diagnosis of the control system, though widely used in the chemical processing industry, is currently lacking in the semiconductor manufacturing industry. This work provides methods for performance assessment of the most commonly used control system in this industry, namely, run-to-run process control. First, an iterative solution method for the calculation of best achievable performance of the widely used run-to-run Exponentially Weighted Moving Average (EWMA) controller is derived. A normalized performance index is then defined based on the best achievable performance. The effect of model mismatch in the process gain and disturbance model parameter, delays, bias changes and nonlinearity in the process is then studied. The utility of the method under manufacturing conditions is tested by analyzing three processes from the semiconductor industry. Missing measurements due to delay are estimated using the disturbance model for the process. A minimum norm estimation method coupled with Tikhonov regularization is developed. Simulations are then carried out to investigate disturbance model mismatch, gain mismatch and different sampling rates. Next, the forward and backward Kalman filter are applied to obtain the missing values and compared with previous examples. Manufacturing data from three processes is then analyzed for different sampling rates. Existing methods are compared with a new method for state estimation in high-mix manufacturing. The new method is based on a random walk model for the context states. This approach is also combined with the recursive equations of the Kalman filter. The method is applied to an industrial exposure process by extending the random walk model into an integrated moving average model and weights used to give preference to the context that is more frequent. Finally, a performance metric is derived for PID controllers, when they are used to control nonlinear processes. Techniques to identify nonlinearity in a process are introduced and polynomial NARX models are proposed to represent a nonlinear process. A performance monitoring technique used for MIMO processes is then applied. Finally, the method is applied to an EWMA control case used before, a P/PI control case from literature and two cases from the semiconductor industry. / text