21793 research outputs found
Sort by
Selection of P3 delivery methods for Sustainable Social Infrastructure Projects Using the Analytical Hierarchical Process
Get PDFThis thesis studies the necessary shift in screening practices of public-private partnerships (P3) projects in Canada, moving beyond traditional qualitative criteria to include broader environmental, social, and governance (ESG) project objectives. The current P3 screening, while effective, needs adaptation to align with Canadian societal and environmental infrastructure goals. In response, this thesis focuses on three objectives aimed at improving social infrastructure P3 procurement and promoting sustainable and responsible project management practices for these projects. Firstly, it identifies and describes Canadian-specific ESG criteria important for ensuring responsible sustainability in delivering social infrastructure projects. Secondly, it develops an ESG-PPP screening matrix to evaluate social infrastructure projects based on responsible sustainability thresholds, determining their suitability for P3 procurement. Thirdly, it implements a multi-criteria analysis using the Analytic Hierarchy Process (AHP) to determine the most appropriate P3 model for social infrastructure projects, considering the identified ESG criteria and quantitative value-for-money criterion. The AHP-PPP selection tool is applied to three case studies analyzing two AHP scales to assess their consistency ratios and the reliability of the P3 selection results. The results indicate that the balanced-n scale exhibit lower inconsistency ratios compared to the fundamental AHP scale, and decisions on P3 options remained consistent across all case studies using both scales, suggesting that the Fundamental AHP scale remains reliable if decision-makers accurately reflect the relative importance of P3 options. Overall, this thesis addresses the increasing need for sustainable and responsible management of social infrastructure projects in Canada by integrating ESG factors into the current procurement process
Development and Implementation of Deep Learning Algorithms for Restoring Images Degraded by JPEG Compression
Get PDFJPEG is one of the most popular image compression techniques in the world. Its effectiveness has led to it being used in diverse sectors such as satellite imaging, medical imaging, image storage systems and multimedia. With the diverse use of JPEG compression algorithms, it has also become necessary to develop deblocking algorithms to mitigate the compression loss caused by the compression. With the advent of deep learning, several
methods have been developed for JPEG image deblocking. The quality factor or QF value is vital to the compression process. Most deep JPEG deblocking networks face the challenge of requiring the image to be compressed by a QF value which is part of the training
process. If the image is compressed by any other QF value, the performance and deblocking quality of the network severely degrades.
In this thesis, two different schemes are proposed to solve this issue. The first proposed scheme aims to tackle the problem from the out-of-distribution point of view, whereas the
second proposed network aims to tackle the problem from a meta-learning point of view. The effectiveness of the proposed schemes is validated by conducting experiments employing two different benchmark datasets. The proposed networks are shown to outperform the state-of-the-art deep JPEG deblocking networks as shown by the quantitative and qualitative comparative studies
Multi-Valued Model Checking IoT and Intelligent Systems with Trust and Commitment Protocols
Get PDFAbstract
Multi-Valued Model Checking IoT and Intelligent Systems with Trust and
Commitment Protocols
Ghalya Alwhishi, Ph.D.
Concordia University, 2024
In the era of connectivity, numerous domains utilize multi-sensor Internet of Things
(IoT) and Intelligent Systems (IS) applications, which involve complex interactions among
numerous components in open environments. This complexity challenges the verification of
these systems’ reliability and efficiency. This study pioneers the verification of IoT applications
and intelligent systems within multi-source data environments, employing multi-agent
commitment and trust protocols, particularly in uncertain and inconsistent settings.
Our research introduces efficient frameworks to model and verify these systems, incorporating
commitment and trust protocols in settings characterized by uncertainty and
inconsistency. We extend existing logics of commitment CTLcc and CTLc and the logic of
trust TCTL to multi-valued cases for reasoning about uncertainty and inconsistency. We
introduce 3v-CTLc and 3v-CTLcc, three-valued logics of commitment for reasoning about
uncertainty, and 4v-CTLc and 4v-CTLcc, four-valued logics of commitment for reasoning
about inconsistency. In the context of trust, we introduce 3v-TCTL and 4v-TCTL, multivalued
logics for reasoning about uncertainty and inconsistency over systems with trust
protocols.
To address the complexity and time needed for developing direct algorithms, coupled
with the scarcity of multi-valued model checking tools, we developed reduction algorithms.
These algorithms transform the introduced multi-valued logics of commitment and trust
into their classical case or into CTL, facilitating interaction with efficient model checkers
such as MCMAS+ and MCMASt, and NuSMV, respectively. To demonstrate the practicality and applicability of the tool in real settings, we presented
and reported experimental results over multiple IoT and IS applications in healthcare,
finance, and smart buildings. Our findings indicate that the proposed approaches and the
MV-Checker tool are highly efficient and scalable, providing accurate results under varying
conditions
A Novel Convolutional Neural Network Pore-Based Fingerprint Recognition System
Get PDFBiometrics play an important role in security measures, such as border control and online transactions, relying on traits like uniqueness and permanence. Among the different biometrics, the fingerprint stands out for their enduring nature and individual uniqueness. Fingerprint recognition systems traditionally rely on ridge patterns (Level 1) and minutiae (Level 2). However, these systems suffer from recognition accuracy with partial fingerprints. Level 3 features, such as pores, offer distinctive attributes crucial for individual identification, particularly with high-resolution acquisition devices. Moreover, the use of convolutional neural networks (CNNs) has significantly improved the accuracy in automatic feature extraction for biometric recognition.
A CNN-based pore fingerprint recognition system consists of two main modules, pore detection and pore feature extraction and matching modules. The first module generates pixel intensity maps to determine the pore centroids, while the second module extracts relevant features of pores to generate pore representations for matching between query and template fingerprints. However, existing CNN architectures lack in generating deep-level discriminative feature and computational efficiency. Moreover, available knowledge on the pores has not been taken into consideration optimally for pore centroids and metrics other than Euclidean distance have not been explored for pore matching.
The objective of this research is to develop a CNN-based pore fingerprint recognition scheme that is capable of providing a low-complexity and high-accuracy performance. The design of the CNN architecture of the two modules aimed at generating features at different hierarchical levels in residual frameworks and fusing them to produce comprehensive sets of discriminative features. Depthwise and depthwise separable convolution operations are judiciously used to keep the complexity of networks low. In the proposed pore centroid part, the knowledge of the variation of the pore characteristics is used. In the proposed pore matching scheme, a composite metric, encompassing the Euclidean distance, angle, and magnitudes difference between the vectors of pore representations, is proposed to measure the similarity between the pores in the query and template images.
Extensive experiments are performed on fingerprint images from the benchmark PolyU High-Resolution-Fingerprint dataset to demonstrate the effectiveness of the various strategies developed and used in the proposed scheme for fingerprint recognition
“We succeeded because we were women”: Revisiting Images from the Early Days of Caedmon Records
Get PDFAs part of The Tape Box series, Maya Schwartz undertakes a close listening to a rare series of photographs that are part of a larger story of audio recordings and of collaboration
Integrating Trauma-Informed Art Therapy into Multimodal Partial Hospitalization Programs for Adolescents with Anorexia Nervosa: A Theoretical Intervention-Based Research Study
Get PDFThis paper seeks to explore the application and integration of trauma-informed and attachment-focused art therapy with adolescents with anorexia nervosa in a partial hospitalization program setting. The primary research question is: How can a trauma-informed art therapy group program be designed to support adolescents with anorexia nervosa within a multidisciplinary partial hospitalization context? This paper will also address the following subsidiary questions: 1) What might a trauma-informed art therapy group in this population and setting look like? and 2) What are the current best practices and barriers to integrating art therapy within multimodal treatment teams for eating disorders in hospital settings? Research questions are explored using a theoretical intervention research methodology to construct problem and program theories that contribute to the development of both a theoretical foundation and a practical understanding of the factors influencing and mediating treatment for this population. Literature from interdisciplinary fields and diverse sources will be examined and analyzed using a narrative synthesis to assess the current needs and inform the development of an 8-week group art therapy intervention employing a trauma-informed and phase-based approach. This exploration sheds light on the intersections of trauma, attachment, anorexia nervosa, and art therapy. Recommendations for further research including future pilot studies to improve findings are highlighted
Cyber-attacks and Their Impact on The Financial Markets: An Empirical Analysis
Get PDFCyber-attacks are a global concern, escalating in frequency. Our study explores whether and how cyber-attacks affect the financial performance of the targeted firms. Unlike previous research, we examine not only attack incidence but also the financial consequences, considering attack and firm characteristics. In addition, our study is the first to explore whether the time frame over which an attack affects a targeted firm varies depending on the type of attack. While the former is reasonably well documented, the latter is not. Market responses are diverse and sometimes delayed, influenced by breach disclosure laws. Our study further examines whether market responses differ when the affected entity is a parent or a subsidiary. Our results suggest that — because each cyber-attack has its own distinct characteristics — an aggregation of instances produces diminished or no results because markets respond within different time frames and in different directions. Our findings disentangle these effects and thus make an important contribution: contrary to the often insignificant findings presented by prior studies, we find that cyber-attacks have, on average, severe financial repercussions for the target firms. Parent firms appear more adept at handling these repercussions and may benefit from their size and diversification, allowing them to mitigate negative stock price effects, while subsidiary firms are hit harder. Finally, we estimate the probability of a cybersecurity incident: we report a convex relationship between a firm’s asset size and its susceptibility to (successful) cyber-attacks and a concave relationship between a firm’s net income and its cybersecurity risk
Development and Testing of a Three-dimensional Deepwater Oil Spill Model (DWOSM) to Predict the Transport and Fate of Subsea Blowouts
Get PDFOffshore oil exploration and production in deep water are associated with environmental risks to marine ecosystems. An effective oil spill response critically relies on understanding the transport and fate of spilled petroleum in complex marine environmental compartments. Oil spill models have been used for decades to help responders make informed decisions by forecasting the movement and fate of released petroleum fluids. However, few existing operational tools could capture the sophisticated behaviors of deep-sea oil spills under extreme ranges of ambient conditions. A deepwater oil spill model (DWOSM) is developed in this study to predict the trajectory and weathering processes of subsea blowouts. This system incorporates droplet size distribution model, buoyant plume model, near- and far-field particle tracking algorithms, and various advanced fate algorithms into three modules: DWOSM-DSD aims to predict the quasi-stationary droplet size distribution resulting from a blowout; DWOSM-Nearfield is to simulate near-field plume dynamics; DWOSM-Farfield forecasts the trajectory and fate of dispersed oil and gas beyond the near-field. Unlike most other oil spill models, DWOSM introduces near-field particle tracking to enable a smooth transition between near-field and far-field. It also takes advantage of thermodynamic modeling to predict the evolution of oil and gas's physicochemical and thermodynamic properties in deep water. In addition to model development, a state-of-the-art stochastic simulation-based risk assessment framework is improved by embedding the DWOSM and a polycyclic aromatic hydrocarbons (PAH)-related risk evaluation index.
The application of DWOSM is demonstrated in three cases. The first study case is a hypothetical oil blowout in 800 m of offshore waters of East Newfoundland, Canada. The DWOSM and its each module are juxtaposed with some established oil spill models. The verification shows that predictions in DWOSM are primarily in line with other model outputs. Different choices in fate algorithms cause a particular discrepancy in simulation results. Multiple spill scenarios are implemented to investigate the impacts of weathering processes on oil fate numerically, which reveals the vital role of natural dispersion in surface oil mitigation under windy conditions. Second, DWOSM is applied to perform a hindcast of the largest offshore oil spill in US history, the Deepwater Horizon (DWH) blowout. Primary model outputs, such as surfaced gas composition and surface oil trajectory, are validated through field observations and relevant modeling efforts. A good performance is presented in most validation results, manifesting the reliable capability of DWOSM to simulate deep-sea spill behaviors. Last, the newly developed model is integrated into a risk assessment framework to evaluate the subsea blowout risk in the offshore area of East Newfoundland. Data mining techniques are used to extract representative met-ocean conditions from long-term hydrodynamic and atmospheric reanalysis products, making computationally demanding stochastic oil spill modeling practicable. A series of deterministic simulations corresponding to each met-ocean condition are conducted to yield regional oil spill hazard and risk mapping. The spill scenarios with and without applying subsea chemical dispersants are formulated to analyze their efficacy on spill mitigation. The results indicate a low-risk level around the nearshore waters of the study area, but PAH exposure can jeopardize the aquatic biota in the oil-infested region. Moreover, dispersant use can reduce the risk peak but facilitate the dissemination of oil spills.
In conclusion, this research contributes a novel modeling toolkit for predicting the complex behaviors of deep-sea blowouts and an improved stochastic simulation-based risk assessment methodology to quantitatively evaluate the regional subsea spill risk
Modularized Directed Greybox Fuzzing for Cross-Architecture Binary Analysis
No full textDirected Greybox Fuzzing (DGF) has proven effective in vulnerability detection, such as bug reproduction and patch testing. Despite this, existing directed fuzzers are often complex monolithic tools that lack modularity and have limited binary support. This constrains their usability on complex software or when the source code is unavailable; a complexity encountered when fuzzing embedded systems. In this thesis, we address these limitations by introducing the Directed Fuzzing Toolkit (DRIFT) as a foundational platform for directed fuzzing within the modular framework of LibAFL. DRIFT modularizes techniques from the state-of-the-art directed fuzzer AFLGo and adapts them for binary applications across CPU architectures. This design augments fuzzers built on top of LibAFL with directed fuzzing capabilities at the binary level thereby enhancing the applicability of DGF on a variety of targets and facilitating the adaptation of these techniques for IoT fuzzing. Our evaluation of DRIFT shows a 90% correlation in distance metric computation over binary over multiple architectures compared to its source-code counterpart. Fuzzing performance was also notable despite operating over emulation. In benchmarks, DRIFT’s performance exceeds the original fuzzer with doubled bug discovery rates and 9-40x faster exploitation times. This accomplishment is attributed to the advantages conferred by our toolkit’s modular design and its native integration with LibAFL. Additionally, DRIFT introduces a profiling platform for directed fuzzing metrics and seamless integration with the Magma benchmark. Together, these features position it as a practical advancement in directed fuzzing within LibAFL
Failure and Reliability Analysis of Composite Pipes for Oil and Gas Drilling Applications
No full textComposite pipes are increasingly replacing traditional materials like steel in various industries due to their high strength-to-weight ratio. This is particularly beneficial for the oil industry, especially in drilling applications where structures face multi-axial loadings.
However, the anisotropic mechanical behavior of these materials adds complexity to their analysis. Moreover, their mechanical properties are statistically variable, affecting component response and failure.
To address these challenges, mathematical models are employed to analyze composite pipe performance, and statistical methods are used to assess their reliability. The primary focus is on developing a composite pipe design capable of withstanding anticipated drilling loads. This involves determining the optimal layup design and
identifying first-ply failure envelopes under combined tension, torsion, and pressure using three-dimensional finite element models in ABAQUS software.
Furthermore, the occurrence of delamination between layers is investigated by integrating the cohesive zone model into simulations. The reliability of composite pipes with suitable layup designs is quantified by considering the statistical variation of
mechanical properties. Reliability-based first-ply failure contours are determined by analyzing the load distribution that each pipe, with specific mechanical properties, could withstand just before damage occurs.
In conclusion, this work aims to develop a comprehensive understanding of composite pipe behavior under drilling conditions. By utilizing advanced modeling techniques and statistical analysis, it seeks to optimize composite pipe designs for enhanced reliability
and performance in the oil industry