University of Bridgeport

Delivery reliability remains a critical determinant of competitiveness in modern supply chains. Despite technological advancements, many organizations continue to experience delays caused by operational and logistics-related inefficiencies. This study investigates the key operational and transportation factors influencing delivery delays using business intelligence and data-driven analytical techniques applied to secondary supply chain data. The research integrates regression modeling, clustering analysis, and mode-based performance evaluation to examine how lead time components and transportation modes jointly affect delivery reliability. The findings demonstrate that delivery delays emerge from systemic interactions among processing, shipment, and coordination phases rather than from isolated bottlenecks. Furthermore, distinct delivery- performance profiles highlight the importance of risk- based order segmentation and mode-specific logistics strategies. The study contributes to the supply chain literature by proposing an integrated analytical framework that combines business intelligence tools with operational modeling to enhance delivery performance management. The results provide actionable insights for managers seeking to improve reliability, optimize lead-time structures, and implement data-driven decision support systems in global supply chain environments.

Abstract Background: Obesity drives a substantial population burden of atherogenic dyslipidemia, contributing to cardiovascular morbidity and mortality. Systemic inflammation is a modifiable pathway linking adiposity to dyslipidemia; however, its population-attributable contribution has not been quantified in high-risk populations. Objective: To estimate the proportion of obesity-related hypertriglyceridemia statistically attributable to systemic inflammation in a national sample of U.S. adults with smoking history and evaluate implications for population health strategies. Methods: Cross-sectional analysis of 863 adults from the Midlife in the United States (MIDUS) Refresher Biomarker Project. We estimated mediation via C-reactive protein (CRP) using a product-of-coefficients approach, adjusting for age, sex, smoking intensity, statin use, and alcohol consumption. Multiple imputation addressed missingness. Robustness was assessed through phantom variable sensitivity testing. Results: In this sample of ever-smokers (mean age 52.7 years, 52.1% female), obesity prevalence (BMI ≥30 kg/m²) was 38.2%, 28.6% had elevated CRP (>3.0 mg/L), and 24.3% had hypertriglyceridemia (>150 mg/dL). CRP mediated 24.7% (95% Monte Carlo CI: 11.9% to 41.3%) of the BMI-triglyceride association. This corresponds to an estimated inflammation-mediated triglyceride increase of 2.7% (95% CI: 1.0% to 4.5%) per 5 kg/m² BMI increment. Sensitivity testing across 27 plausible confounding scenarios showed consistent mediation. Conclusions: Systemic inflammation accounts for approximately one-quarter of obesity-related dyslipidemia burden in this national sample of adults with smoking history. These findings quantify a substantial modifiable pathway and support integrated public health strategies that address both adiposity and inflammation, particularly in high-risk populations. CRP may have utility in identifying high-risk individuals for targeted intervention.

Abstract In today’s digital environment, social media platforms offer many free services that have become integral to users’ daily lives. However, there are serious privacy concerns associated with this convenience, especially given how artificial intelligence (AI) is being incorporated into user data management. This study explores the social and ethical dimensions of the trade-off between the benefits of free platform use and the privacy concerns emerging from AI-powered targeted advertising. Users are generally more concerned about how AI algorithms use their data for advertising and personalization than they are about the collection of their data. AI technologies have significant issues with surveillance, behavioral profiling, and data commodification, even though they improve user experience through personalization. The study presented is thorough literature-based research that explores user knowledge gaps and perceptions of risk and awareness of AI-based features in the use of social media. It has been discovered that there has been an increase in attention to data collection, but a lack of knowledge about the AI mechanisms, thus leading to an overall problem of the so-called privacy paradox, in which continued use is promoted despite the lack of trust in it. The study suggests the proposed framework model of the study to guide the evaluation of the user attitudes and intentions. These lessons are meant to help social media companies; policy initiators and digital rights activists understand how to strike a balance between technological advances and strong privacy safeguards. Study phases that will be applied later will consist of a Delphi-based qualitative analysis of the internet privacy experts.

This study introduces the Skills Self-Efficacy Scale for Adolescents (SSES-A), a validated tool to measure students’ self-efficacy in five vital 21st-century competencies: problem-solving, communication, creativity, collaboration, and critical thinking. It addresses a critical gap in educational assessment, where traditional approaches have yet to fully integrate and explore these transversal skills within adolescent populations. These competencies are universally recognized pillars essential to thriving in today’s knowledge-rich, globalized society. The SSES-A empowers educators, curriculum designers, and policymakers with a dependable tool to evaluate and gather insights into students’ strengths and weaknesses, guiding targeted improvements to assist in preparing them for diverse global and professional contexts. The dataset comprises responses from 385 high school students across four Connecticut public schools for the deployment validation sample, supplemented by qualitative data from focus groups (23 students), industry professional surveys (30 participants), expert reviews (35 professionals), and cognitive pretesting (40 high school students). The final 21-item scale, refined through rigorous exploratory factor analysis, exhibits robust psychometric properties with a five-factor structure aligned with the targeted skill domains, excellent internal consistency (Cronbach’s alpha = 0.935), and strong content validity (S-CVI = 0.921). This comprehensive dataset enables researchers and educators to profile adolescents’ self-efficacy across key competencies, identify intervention needs, and inform educational strategies that foster academic and professional success. The SSES-A offers a psychometrically sound instrument to assess adolescents’ self-efficacy in essential 21st-century skills.

A 60-year-old male with lifestyle factors including excessive cigarette smoking, high intake of sugared energy drinks, high processed Standard American Diet (SAD), moderate alcohol intake, and lack of exercise presented symptoms of fatigue, frequent urination, poor wound healing, lethargy, insomnia, excessive snoring, and dizziness. A diagnosis of sleep apnea, polyurea, metabolic syndrome, hypercholesterolemia, hypertriglyceridemia, and hyperglycemia was made by the treating physician. The patient was advised a personalized high-quality supplement regimen and diet intervention for six months. The supplement regimen consisted of Designs for Health Red Yeast Rice, Arterosil® HP, Metabolic Synergy™, Berberine-Evail™, GlucoSupreme™ Herbal, and Mito-NR™. The protocol resulted in complete reversal of hypertriglyceridemia, hypercholesterolemia, and hyperlipidemia, significant weight loss, and positive changes in hyperglycemia. This case report investigated the scientific basis of the intervention (nutraceuticals and their constituents and the dietary protocol), rationale, and outcome. case report, metabolic syndrome, hyperlipidemia, hypercholesterolemia, hypertriglyceridemia, type 2 diabetes mellitus.

Non-traditional learners comprise approximately 73% of undergraduate enrollment, representing diverse populations including first-generation college students, adult learners, veterans, multilingual learners, and students with family responsibilities. Despite their numerical dominance, these students face systemic barriers that traditional pedagogical approaches often fail to address. This mixed-methods study examined how Universal Design for Learning (UDL) principles impact non-traditional learners’ educational experiences in higher education. Using a convergent parallel design with 154 participants from a Hispanic-serving institution, the study collected quantitative data through the validated Personalized Learning Supporting Instrument (PLSI) and qualitative data from open-ended questions. The refined 12-item PLSI demonstrated strong psychometric properties. While UDL factors showed limited direct association with GPA overall, Flexible Instructional Methods and Materials significantly predicted academic performance. Qualitative analysis identified six barrier themes (online learning difficulties, course content issues, financial constraints, balancing responsibilities, accessibility challenges, and health interruptions) and five positive impact themes (interactive learning, supportive environments, skill development, goal clarification, and effective assignments). Demographic analysis revealed counterintuitive patterns—students with traditional “barriers” achieved high GPAs at rates of 73–76%, while first-generation students showed the lowest high GPA rate (53.2%). These findings challenge deficit-based assumptions about non-traditional learners while revealing important equity gaps. This study demonstrates both the promise and limitations of UDL for diverse populations, suggesting institutions need comprehensive approaches with differentiated support strategies.

The contemporary financial services landscape faces unprecedented challenges in risk assessment, as traditional methodologies prove inadequate against sophisticated threats while regulatory frameworks demand transparency in automated decision-making processes. The proliferation of digital payment systems has exponentially increased transaction volumes, creating environments where manual risk assessment becomes practically unfeasible. Machine learning technologies offer transformative solutions through advanced pattern recognition and anomaly detection capabilities that exceed conventional rule-based systems. However, current artificial intelligence implementations often operate within silos, creating vulnerabilities in model performance monitoring and regulatory compliance verification. Financial institutions frequently encounter situations where high-performing models fail to meet explainability requirements while transparent systems sacrifice predictive accuracy. A comprehensive Quality Assurance-integrated decision framework addresses these challenges by harmonizing advanced machine learning capabilities with regulatory compliance requirements through real-time decision engines, synthetic data testing protocols, model drift monitoring systems, and explainable AI components. The framework encompasses fraud detection, credit scoring, and continuous monitoring systems designed to operate within existing regulatory constraints while maintaining operational excellence. Implementation demonstrates substantial improvements in detection accuracy, processing efficiency, and operational cost reduction across multiple evaluation dimensions. The architectural approach integrates anomaly detection capabilities, automated quality assurance protocols, and comprehensive audit mechanisms to ensure sustained compliance across diverse operational environments, establishing foundational principles for compliant artificial intelligence deployment in financial technology applications.

This study presents a novel partial differential equation (PDE)-based model for neural stem cell (NSC) therapy in spinal cord injury (SCI) treatment. Existing models often fail to accurately represent NSC migration, differentiation, and interaction with biomaterials and immune responses. By integrating scaffold-mediated diffusion and immune response dynamics, the proposed model offers a more realistic framework for regenerative therapy optimization. The model improves upon previous approaches by providing precise control overgrowth factor distribution, ensuring sustained bioavailability for enhanced NSC survival and functional integration. Applications of this framework extend beyond SCI treatment, with potential implications for neurodegenerative diseases, brain injury repair, and personalized medicine. Future research will focus on experimental validation and machine learning-driven optimization to refine the predictive capabilities of the model.

In the face of increasing natural disasters, ensuring rapid patient transportation by Emergency Medical Services (EMS) is critical to saving lives. The Ambulance Dispatching and Relocation Problem (ADRP) poses a significant challenge, requiring swift allocation of limited ambulance resources. To address this issue, we propose a GPU-Accelerated Slime Mould Algorithm (GPU-SMA) designed for real-time decision-making in disaster scenarios. Our approach leverages the parallel processing power of GPUs using the Compute Unified Device Architecture (CUDA). This significantly reduces computational time, enabling faster and more effective optimization. Additionally, we introduce a new relocation policy that utilizes real-time ambulance data to maintain optimal ambulance positioning. Our method has been using real-world data from the COVID-19 pandemic in Chicago. The results show a remarkable 23x speedup on GeForce RTX 3090 and RTX A4000 GPUs compared to the serial implementation. GPU-SMA outperforms five leading parallel algorithms (GPU-PSO, GPU-APSO, GPU-HHO, GPU-FA, and GPU-BA) in efficiency and effectiveness. A Friedman test confirms the statistical significance of these results.