eScholar

eScholar stores, preserves and disseminates digital copies of the research and scholarly output of eScholar faculty, researchers and students. These can include the following items:

  • Monographs
  • Pre- and post-prints of academic journal articles
  • Theses and dissertations
  • Major projects and papers
  • Reports/working papers and conference proceedings

Materials in eScholar are openly available to the world and discoverable through search engines such as Google Scholar. This high visibility, discoverability, and exposure can lead to increased citation. Contact Library Publishing for more information.

Communities in eScholar

Select a community to browse its collections.

Now showing 1 - 5 of 9

Recent Submissions

  • Thumbnail Image
    Item type: Item , Access status: Open Access ,
    How is fairness measured in practice? A survey and critique of fairness analyses applied to predictive algorithms
    (2026-01-27) Putman, Andrew; Lewis, Peter; Rudoler, David
    The increasingly rapid pace of research and development of prediction algorithms over the past decade has spurred their adoption across seemingly every industry and aspect of life. This growth and the subsequent increase in public awareness have brought a range of concerns into focus regarding the potential bias and fairness impacts of these algorithms. This cross-disciplinary survey assesses peer-reviewed publications from the last 10 years which assessed fairness in a predictive algorithm using real world data. We begin with a descriptive overview of the fields of research, types of publication, and the types of outcomes being predicted. We then engage in a more detailed exploration of the fairness analyses that were applied, the metrics used, and the evidence provided to justify these selections. We also discuss the contexts in which these analyses were conducted. We conclude by highlighting the lack of inclusion of perspectives from people who are impacted by the algorithms being analyzed and discuss directions for future work in this area.
  • Thumbnail Image
    Item type: Item , Access status: Open Access ,
    Design and development of a LIVE Digital Twin methodology for predictive maintenance of bearings in rotary machine systems
    (2025-12-01) Zonta, Tristan J.A.; Barari, Ahmad
    Digital Twin (DT) is a prominent focus for many predictive and prescriptive maintenance strategies. In maintenance, DT is used for connecting the physical and digital models of a maintenance monitoring system and proactively prescribing maintenance solutions to extend the products life. Many of the failures in modern DT can be attributed to the lack of defined structure, the unavailability of failure data to calibrate the systems, and poor connectivity between the physical and digital systems. LIVE provides a systematic approach to implement a DT system in 4 stages of Learn, Identify, Verify and Extend. This thesis uses LIVE DT for dynamic rotary systems connecting the physical asset to its DT to predict failure. In addition, this thesis covers the development of a device that will allow for emulating bearing defects in a controllable and repeatable way for calibrating virtual systems when historical or failure data is unavailable.
  • Thumbnail Image
    Item type: Item , Access status: Open Access ,
    Return on investment in nuclear security: regulations and physical security
    (2025-12-01) Vucicevic, Jelena; Waller, Ed
    Nuclear security is a crucial part of the nuclear industry, and therefore it is important to observe, research and improve its return on investment (ROI). Investments are made not only in the physical security itself, but also in the regulatory framework, and they should be studied holistically when discussing ROI in nuclear security. This research will analyze the regulatory aspect by analyzing administrative monetary penalties (AMP) in the nuclear industry, and the application of drones in nuclear security, from the side of nuclear facility/licensee security. An AMP is a penalty imposed by the regulator, the Canadian Nuclear Safety Commission, without court involvement, in the case of a violation of a regulatory requirement. The research includes analysis of AMP, with the focus on the advantages and disadvantages of the AMP system and recommendations for improvements for future implementation. Furthermore, the research focusses on how AMP are implemented in various low probability-high consequence industries, as well as the nuclear industry in Canada and the United States. The emphasis is on the differences in these industries’ approaches and how they may be used to improve the implementation in the nuclear industry. Nuclear facilities invest significantly in security, both for security guards and security tools. Even though drones are relatively new technology, they have been utilized as a tool in emergency situations, fires, floods, search and rescue operations, and surveillance. Due to their characteristics, they show potential for utilization in nuclear security, as they show certain advantages over traditional security systems. The research describes how drones can be effectively utilized in nuclear power plants to complement existing measures and in some cases, replace them. Finally, through both aspects, it will be shown that ROI in nuclear security can be improved by minor modification of policies or by introducing new tools. Positive ROI is not aways represented by monetary value, but an outcome that meets or exceeds the expectations.
  • Thumbnail Image
    Item type: Item , Access status: Open Access ,
    Long-term spatiotemporal water quality and phytoplankton dynamics in Severn Sound, Ontario: a delisted great lakes area of concern
    (2025-12-01) Tyner, Alana; Kirkwood, Andrea
    Severn Sound is an embayment system located in south-eastern Lake Huron that was previously designated an Area of Concern (AOC) in 1987 as part of the Great Lakes Water Quality Agreement. The AOC designation was largely due to eutrophication, which has been a historical issue in Severn Sound extensively documented by a water quality and phytoplankton monitoring program since 1973. My thesis research analyzed this long-term dataset representing four embayments within Severn Sound as well as nearshore-offshore water quality in the embayments. I found statistically significant differences across stations in water quality parameters such as total phosphorus, total nitrogen, Secchi disk visibility, and water temperature. Overall decreases in total phosphorus and chlorophyll a and increases in Secchi disk visibility illustrate the influences of nutrient reduction strategies, and potential influences from invasive dreissenid mussels. Although an important driver of water quality changes, water temperature change was variable during the study, likely due to site-specific differences and inherent data variability. Generalized Additive Mixed Models showed decreases in total phytoplankton biovolume and different significant drivers across seven major phytoplankton divisions. Permutational Multivariate Analysis of Variance (PERMANOVA) revealed that station, along with period (before, during, and after delisting), and interaction of station and period were significant drivers of phytoplankton community composition. Using redundancy analysis, I found cyanobacterial community assemblages were significantly different both before, during, and after AOC listing, and before and after the zebra mussel invasion, as well as across stations, with period being the highest explanatory variable of community shifts. Generalized Additive Models revealed a significant difference in environmental drivers of three major cyanobacterial genera, Aphanizomenon, Dolichospermum, and Microcystis. Overall findings indicate that cyanobacterial communities shifted in response to nutrient mitigation, climate, and the zebra mussel invasion. In addition, I found statistically significant differences in nearshore-offshore chlorophyll a, conductivity, total ammonia/ammonium nitrogen, total organic nitrogen, and total phosphorus. A PERMANOVA also revealed that nearshore-offshore water quality was significantly different across region but found no statistical difference in embayments. Overall, these findings can be used to inform management decisions in the Severn Sound region, other Areas of Concern, and regions experiencing eutrophication.
  • Thumbnail Image
    Item type: Item , Access status: Open Access ,
    Network disaster recovery via dynamic service-aware risk management
    (2025-11-01) Taghavi Motlagh, Sara; Heydari, Shahram Shah
    Communication networks are increasingly exposed to natural and human-induced disasters, which result in service disruption and economic losses. Most existing disaster recovery frameworks treat risk as a static factor, overlooking the fact that hazard conditions evolve over time. For example, in earthquakes, failures propagate outward from the epicenter across successive time steps, creating time-varying and correlated risk for network links. In practice, routing must therefore be both dynamic and service-aware, reflecting class priorities, (Service Level Agreement) SLA targets, and capacity constraints, yet existing models rarely integrate these elements in a unified optimization framework. This thesis formulates dynamic, service-aware routing under disaster conditions as a multi-objective mixed-integer nonlinear program (MINLP). The model jointly optimizes five objectives: maximizing revenue and survivability, while minimizing service-aware risk, overutilization cost, and SLA penalties, subject to capacity and service-level constraints. A dynamic risk model is introduced that considers time-varying link failure probabilities and is weighted by service priorities. This includes both hazard progress and the importance of impacted services. SLA compliance is preserved by class of service availability constraints, which guarantee the preservation of high-priority traffic during disruptions. We solve the model using the Non-dominated Sorting Genetic Algorithm II (NSGA-II), which approximates the Pareto front and provides a decision-support layer for service providers. Experiments are conducted on real-world network topologies (ERnet, France, Giul39) under staged and dynamic failure scenarios. The results show that adaptive routing shifts flows from high-risk to safer links as hazards evolve, improving survivability and reducing SLA penalties while explicitly exposing trade-offs with revenue and overutilization. Sensitivity analyses show how link capacity, service request volume, and failure probability affect the balance among the five objectives. By embedding dynamic risk, service differentiation, and capacity constraints in a single multi-objective formulation, this work enables decision-makers to explore balanced recovery strategies rather than commit to a single prescriptive solution. The approach provides a practical tool for evaluating disaster-resilient routing policies under uncertainty, which helps strengthen communication networks against evolving natural and human-induced disruptions.