Search

Plenty of literature has examined the vulnerability of transportation networks. To identify appropriate measures of connectivity for heavy rail systems, this research presents a comprehensive measure named Degree of Nodal Connection (DNC) index along with a new classification of transfer stations – Mandatory Transfer (MT), non-Mandatory Transfer (non-MT), and End Transfer (ET). The DNC index reevaluates nodal connectivity among various types of transfer stations in heavy rail networks with multiple lines. The concept of partial node failure is proposed and addressed in network modeling, and the disruption results are compared between partial and complete node failures using four local and global DNC indexes. Numerical assessment is presented with a case study of major heavy rail networks in the United States. To incorporate network flow in addition to nodes and links, the research proposes two network optimization models in order to identify and evaluate critical components of a flow-based network from the perspective of shortest paths. The Shortest Path Network model (SPN) identifies the optimal flow distribution across the network where all flows find their shortest paths from origin to destination. The Assessing Nodal Disruption in Shortest Paths Network model (AND-SPN) assesses the influence of r nodes’ disruption on the network flow pattern. Supplementary criticality indicators are provided to reflect average arc use, average arc cost, and average arc flow. In the case study of the Amtrak rail system, the criticality of stations is evaluated comprehensively in terms of objective function and criticality. To accommodate partial node failure, the SPN model is further expanded to the SPN for Partial node failure model (SPNPr) by introducing link attribute index and distance update constraints. A case study on the Washington Metropolitan Area Transit Authority (WMATA) network is carried out using daily and monthly station-to-station flow to assess nodal criticality of MTs and thus network vulnerability. From different perspectives of concern, the indexes and optimization models proposed in the dissertation can help decision makers and planners to decide which rail stations are the most important to protect during special events or disasters or when seeking to reduce transportation network vulnerability.

Food insecurity has been associated with overweight/obesity in U.S. women. Several hypotheses have been proposed to address this paradoxical association, but none has yet been tested. This dissertation is designed to test the “monthly cycle of food abundance and food shortage” hypothesis, and to examine the effects of food stamp program (FSP) participation, disordered eating, and dietary intake patterns on the association. It is hypothesized that food insecure women would experience a monthly cycle with higher total energy intake (TEI) and household food stores at the beginning of the month, followed by a more limited TEI and food supply at the end. The dissertation compared food insecure and overweight/obese (FIS/ovob) women with three other women groups: food secure and normal weight (FS/norm), food secure and overweight/obese (FS/ovob), and food insecure and normal weight (FIS/norm). The monthly variations in TEI and food stores were assessed in a sample of low-income women in Ohio, by comparing the energy intake from the first ten days with that of the last ten days of the month during three continuous months. For FIS/ovob women, significant decreases were found in the total number of food items (Month 1: 87.74 vs. 68.26, Month 2: 83.3 vs. 72.2, Month 3: 88.81 vs. 75.3, p

A class of Orthogonal Latin Hypercubes (OLH) which preserves the orthogonality among columns is constructed. Applying an Orthogonal Latin Hypercube design in a computer experiment benefits the data analysis in two ways. First, it retains the orthogonality of traditional experimental designs. The estimates of linear effects of all factors are not only uncorrelated with each other, but also uncorrelated with the estimates of all quadratic effects and bi-linear interactions. Second, it can facilitate non-parametric fitting procedures, since a good space filling design within the class of Orthogonal Latin Hypercubes can be selected. Furthermore, the structure of Orthogonal Latin Hypercubes is established in the vector space $\{0,1\}\sp{m}.$ An upper bound on the number of columns of an OLH is given. OLHs whose number of columns reaches the upper bound, are found for $m=3,4$ while the cases for $m>5$ are still unknown. A computer experiment using an OLH design conducted at Ford Motor Co. which actually motivated this research is also presented. Optimal Latin hypercube designs are widely used in computer experiments. However, existing search methods are slow, especially for Latin hypercubes with large dimensions. A class of Symmetric Latin Hypercube which has good geometric properties is proposed. The CP algorithm due to Li and Wu (Technometric, 1997) is used to search for optimal Symmetric Latin Hypercubes. Compared with the search in the family of Latin hypercubes using the CP algorithm, its search time is drastically reduced and the Latin hypercubes that can be realistically obtained are likely to be better. In addition, the structure of Symmetric Latin Hypercube provides good properties for fitting a polynomial model.