Theses and Dissertations

Spring 2017

Shen, Wenhan

and parameter uncertainties. Robust optimization (RO) approximation, a novel method dealing with joint chance constraints, is investigated to solve CCMPC problem. This method leads to results close to the true optimal and is not restricted to certain types of distribution. This work is further applied

on the steam assisted gravity drainage (SAGD) process. Constraint violations are greatly reduced by using the RO method. For system noises, the RO method can be directly applied with the inclusion of uncertainty sets. The type of uncertainty set is selected based on the distribution. Two-layer

optimization is proposed, one layer guarantees probability satisfaction and the other layer deals with optimizing the cost. Compared with traditional analytical methods, RO method is not limited to specific distribution and shows better performance in objective function. For parameter uncertainties, random

Fall 2010

Tremblay, Marie Anne

Urbanization is viewed as a major threat to global biodiversity because of its role in the loss and fragmentation of low-lying, productive habitats associated with coastal plains and river valleys. My study examines the effects of urbanization on the movements and distribution of songbirds in

, and isolation from natural features on the distribution of songbirds. In 563 playback trials involving the responses of 2241 birds, I found that the size of the gap in vegetation was the most important determinant of movement across linear features; the likelihood of movement sharply decreasing as

surveys revealed that natural forest stands played a critical role in sustaining regional avian diversity in the study area. Moreover, functional distance to the nearest forested natural area or water body often explained more variation in the probability of occurrence of focal species than straight-line

Fall 2015

Zhu, PeiCan

simulation. Studies of a simple p53-Mdm2 network reveal that random gene perturbation has a greater effect on the steady state distribution (SSD) compared to context switching activities. Secondly, stochastic multiple-valued networks (SMNs) are investigated to evaluate the effect of noise in a WNT5A network

Originally proposed in the 1960s, stochastic computation uses random binary bit streams to encode signal probabilities. Stochastic computation enables the implementation of basic arithmetic functions using simple logic elements. Here, the application of stochastic computation is extended to the

domain of gene network models and the fault-tree analysis of system reliability. Initially, context-sensitive stochastic Boolean networks (CSSBNs) are developed to model the effect of context sensitivity in a genetic network. A CSSBN allows for a tunable tradeoff between accuracy and efficiency in a

Spring 2016

Bai, Guanghan

the probability of successfully sending a required amount of flow, d, from node s to node t, which is the probability that the flow throughput is not less than d. The capacity (state) of each component can take discrete, non-negative integer values from 0 to its maximum capacity, following a certain

probability distribution. The overall objective of multistate network reliability is to provide engineers and managers useful tools to enhance their ability for design and maintenance of such networks. However, despite the increasing complexity of modern networks, the size of the network that can be analyzed

focus on one specific demand at a time. However, during the design phase or operation phase, we are often interested in system reliability with respect to multiple possible demand levels, in order to obtain a complete picture of the system capability. Thus, an efficient and systematic method is

Spring 2017

Fallah Nafari, Saeideh

framework was also developed to estimate the probability distributions of maximum tensile and compressive bending stresses in the rail head and base, which are necessary for calculating the rail reliability under applied loading. The accuracy of the proposed equations was first verified using a numerical

method was developed which allowed the simulation of a stochastically varying track modulus along the track. Track models with different track modulus distributions were developed and the resulting VTD and rail bending stresses under moving wheel loads were calculated. The mathematical correlations

between the inputted track modulus, modelled VTD and rail bending stresses were quantified using statistical approaches. Based on the results, equations were proposed to estimate the statistical properties of track modulus and rail bending stresses over track windows using the VTD measurements. A

Fall 2016

Wu, Yifan

In an online learning problem a player makes decisions in a sequential manner. In each round, the player receives some reward that depends on his action and an outcome generated by the environment while some feedback information about the outcome is revealed. The goal of the player can be various

few probes as possible. Then we study the side observation model in the regret minimization scenario. We derive a novel finite time distribution dependent lower bound and design asymptotically optimal and minimax optimal algorithms. Last we investigate the conservative bandit problem where the

. In this thesis we investigate several variants of online learning problems with different feedback models and objectives. First we consider the pure exploration problem with multi-action probes. We design algorithms that can find the best one or several actions with high probability while using as

Fall 2015

Lajevardi, Saina

. Notable features of this thesis are (1) addressing information loss in facies upscaling process through a proposed measure which captures variability on non-major facies; (2) proposing a novel inverse modeling approach to estimate shale continuity in the form of a probability distribution function; (3

Fall 2015

Moreira, Dario A.

significantly affected by native forest, road density and the presence of dogs (Canis familiaris). The magnitude of these effects were also influenced by the time of day and spatial scale. The positive effect of native forest on occupancy probability was stronger during the night for the Darwin's fox

Mammalian carnivores inhabiting human-dominated landscapes may face reduced and heterogeneous distributions of feeding resources while being more exposed to humans and introduced carnivores. Therefore, sustainable landscape planning intended to conserve carnivores in human-dominated landscapes

requires the use of a multi-dimensional approach that integrates different conceptual and methodological components, such as: 1) habitat and prey selection patterns of carnivores at different spatio-temporal scales; 2) habitat selection models including fine-grain information of habitat structure; and 3

Fall 2017

Al-Haji, Ahmad

Trading patterns in US financial markets have undergone significant changes in the past two decades. Using a 21-year (1993-2013) sample of intraday data, this thesis documents the ways in which the size distribution of trades—that is, the distribution of trades based on their dollar value—has

trades. The results of my test point in the direction of this conjecture. I also test whether a temporary increase in information-based trading shifts the distribution of trades toward smaller transactions. I classify stocks according to their probability of information-based trading (PIN) values during

changed over this period and examines changes in the price impact of trades and activities of informed traders. Chapter 1 examines changes in trading activity and quantifies changes in the size distribution of trades between 1993 and 2013. On average, the daily trading volume per stock increased from

Spring 2017

Hunt, Anjolene R

found that use of post-harvest stands did not affect probability of pairing or fledging young, but that pairing success was lower when male densities were high. My final objective was to discuss potential reasons for discrepancies between conclusions about the effects of forestry on Canada Warblers

Recovery strategies for species at risk are legally mandated in Canada and the Government of Canada must identify which habitat is important for a species and which activities result in its destruction. The Canada Warbler (Cardellina canadensis) has been designated as a threatened species in Canada

due to large population declines (~3% annually over the last 50 years). Forestry has been identified as a threat, but some studies suggest it can create productive breeding habitat. I quantified multiple orders of habitat use to study the response of the Canada Warbler to forestry, accounted for the