Thesis (Ph. D.)--University of Texas at Austin, 2001. / Vita. Includes bibliographical references. Available also from UMI/Dissertation Abstracts International.

Not available / text

Thesis (Ph. D.)--University of Wisconsin--Madison, 1958. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 147-149).

Thesis (M.S.)--Ohio University, November, 2002. / Title from PDF t.p. Includes bibliographical references (leaves 116-117).

Bus bunching occurs when two or more buses travel head to tail. It is an annoying problem in public transportation because it increases passengers' average waiting time and traveling time, wastes bus capacity, reduces the frequency of bus service and increases the pressure on bus drivers. So eliminating bus bunching is important in public transportation. Eliminating bus bunching is highly challenging due to the complexity and variability of the bus dynamics. Bus bunching results from a positive feedback mechanism of headway evolution, which is a flaw born with the bus system. In this thesis, we quantify the intensity of the tendency to bus bunching and propose a headway control modeling framework to reverse tendency. Our framework subsumes many headway control schemes to coordinate buses and so enables batch analysis. Given different headway information, our framework produces different control schemes under which headways self-equalize. The stability of the bus system under control is characterized by a single measure and it can be optimized. Besides, the bus system under control is robust against traffic conditions and the level of ridership. The framework is based on a snapshot model capturing the bus dynamics including the tendency to bunch by taking traffic conditions and the level of ridership into account. It is linear and time-invariant, which makes the bus dynamics tractable. This model considers a single control point and constant bus velocity in a deterministic manner, but it can be extended to handle many control points, inhomogeneous velocity along the route, and randomness. Using our framework, we further study two simple control schemes---Threshold control and ``Prefol". Threshold control drives headways to self-equalize the fastest but the corresponding bus system needs large slack time for robustness. "Prefol" needs small slack time but headways self-equalize slower. We hybridize them and find the hybrid control scheme balances robustness and fast headway equalization. We also show that it outperforms several state-of-the-art control schemes in tests on a simulated bus route in Chicago.

Thesis (M.S.)--University of Missouri--Rolla, 2007. / Vita. The entire thesis text is included in file. Title from title screen of thesis/dissertation PDF file (viewed December 4, 2007) Includes bibliographical references (p. 67-69).

Thesis (Ph. D.)--University of Chicago, Dept. of Economics, 2002. / Includes bibliographical references. Also available on the Internet.

Thesis (M. Phil.)--University of Hong Kong, 1983.

published_or_final_version / Mathematics / Master / Master of Philosophy

Self-Organizing Maps (SOM) can provide a visualization for multi-dimensional data with two dimensional mappings. By applying unsupervised learning techniques to SOM representations, we can further enhance visual inspection for change detection. In order to obtain a more accurate measurement for the changes of self-organizing maps beyond simple visual inspection, we introduce the Gaussian Mixture Model (GMM) and Kullback-Leibler Divergence (KLD) on top of SOM trained maps. The main contribution in this dissertation focuses on adding numerical methods to SOM algorithms, with anomaly detection as example domain. Through extensive traced-based simulations, it is observed that our techniques can uncover anomalies with an accuracy of 100% at an anomaly mixture-rate as low as 12% from the CTU-13 dataset. Tuning of the KLD threshold further reduces the mixture-rate to 7%, significantly augmenting visual inspection to assist in detecting low-rate anomalies. Suitable hierarchical and distributed SOM-based approaches are also explored, along with other approaches in the literature. Hierarchies in SOM can show the correlations among the neural cells on the self-organizing maps. In order to obtain a higher accuracy for anomaly detection, a new dimension of labels is suggested to be added in the second layer of SOM training. Also for more general distributed SOM-based algorithms, we investigate the use of principal component analysis (PCA) for the separation of dimensions. With the transformed dataset from PCA, the inner dependencies can be reserved in a manageable scale. As a case study, this dissertation uses a SOM-based approach for anomaly detection in Supervisory Control And Data Acquisition (SCADA) networks. We further investigate the use of SOM for the Quality of Service (QoS) in the scenario of wireless SCADA networks. Solving the problem of long computing time of optimizing the cached contents, the new SOM-based approach can also learn and predict the sub-optimal locations for the caching while maintaining a prediction error of 28%. / Graduate