First-year international students face significant social challenges, including potential reductions in social support and difficulties in forming friendships. A nuanced examination of these social challenges is important because they may contribute to experiences of loneliness and social isolation, and in turn, have a negative impact on mental health and academic performance. While quantitative research has established the prevalence, predictors, and outcomes of international students’ loneliness, a qualitative approach can provide a richer, more in-depth account of their lived experiences. As such, we interviewed 15 first-year international students enrolled at a Canadian university about their loneliness and social isolation, the development and maintenance of their social networks, and the challenges they experienced when adjusting to a new cultural and academic environment. Key themes included the emotional and psychological toll of loneliness, especially during holidays, the benefits of solitude, difficulties in forming new friendships in Canada, the lack of opportunities to interact with classmates outside the classroom setting, as well as support from family, friends, and romantic partners from Canada or their heritage culture. Some participants reported feeling a lack of belonging after attending social events, realizing they did not share the same values as others. Family support (e.g., through video calling) was an important source of social support despite the lack of in-person interactions. These results may help to inform university initiatives for enhancing social support systems and assisting international students’ transition to life in Canada’s evolving sociopolitical landscape.
Air cargo transportation plays a key role in global trade, especially for time-sensitive and high-value products. Accurate prediction of air cargo demand is essential for informed decision-making on infrastructure planning, capacity management, and resource allocation across the air transport sector. While previous studies have largely advanced air cargo demand forecasting through model development and performance comparison, this study introduces a framework to explain and compare established predictive models based on data characteristic analysis (DCA). By examining key time series characteristics, such as stationarity, seasonality, and complexity, across statistical, machine learning, and deep learning approaches, this research examines how intrinsic properties of demand data are associated with forecasting performance. A rolling horizon design is also employed to evaluate how dynamic changes in data characteristics influence model performance over time. The findings reveal that statistical models are particularly sensitive to the mutability and complexity of air cargo demand data, whereas machine learning and deep learning models demonstrate stronger adaptability under diverse demand data conditions. Overall, this study shifts the emphasis from developing new forecasting models to explaining model suitability and supporting model selection, offering both theoretical insights and practical guidance for stakeholders by highlighting which models are best suited under specific data conditions.
Zero Energy Deuterium (ZED-2) reactor experiments at the Canadian Nuclear Laboratories have the potential to support the development of advanced power reactors. Relevant experimental data from ZED-2 could be used to validate code predictions of important quantities such as the reactivity temperature coefficient (RTC). However, ZED-2, was originally designed to support CANDU reactor applications, and its applicability to alternate designs is not well established. This work provides a thorough example of how a proposed experiment can be assessed based on its potential to reduce application response uncertainty within the TSUNAMI validation framework. A similar approach could be adopted by reactor designers and experimentalists to explore and optimize future benchmark experiments for novel applications. This work considers two potential ZED-2 experiments for validating predictions of molten salt reactor (MSR) RTCs. The results indicate that valuable experimental data can be gathered from appropriately designed ZED-2 experiments. However, configurations should minimize reliance on driver fuels as they significantly reduce the value of the experimental data.
Hydrothermal carbonization (HTC) converts biomass into carbon-rich material, but improving surface chemistry, fuel quality and combustion properties at reduced severity remains challenging. This study used response surface methodology to determine the extent to which citric acid (CA) catalysis replaced HTC severity in spruce-pine-fir sawdust. Quadratic models for solid yield (SY), higher heating value (HHV), carbon content (CC) and fuel ratio (FR) were statistically significant (R2 > 0.85; p < 0.0001). Numerical CC optimization identified 280 °C, 129 min and 13.9 wt% CA as the optimum condition, where predicted CC, HHV and SY (74.20 wt%, 31.60 MJ kg−1, 52.37 wt%) matched the experimental values (71.02 wt%, 30.24 MJ kg−1, 51.00 wt%) within 2.69-4.50% error. At the milder condition of 230 °C, 135 min and 17.5 wt% CA, hydrochar exceeded the CC and fixed carbon of its non-catalytic counterpart (68.84 vs 63.12 wt%, and 44.04 vs 32.20 wt%, p < 0.001). Relative to non-catalytic HTC at 280 °C, the catalytic 230 °C hydrochar showed 3.5-fold higher BET surface area (87.77 vs 25.08 m2 g−1), higher surface acidity (2.30 vs 1.68 mmol g−1), lower pH, and pHPZC (3.13 vs 3.66; 3.87 vs 4.12). This acidity advantage survived washing, indicating retained structural rather than leachable functionality relevant to contaminant binding. Despite a 50 °C drop in HTC temperature, TGA/DTG showed that CA extended burnout temperature by 86.8 °C (872.9 vs 786.1 °C), while lowering ignition temperature by 79 °C and combustibility index by 11%. CA catalysis thus reduces thermal severity while advancing energy density, surface functionality, combustion profile, and thermal resistance for combined energy-environmental application.
We propose here a novel bivariate gamma degradation model to characterize products that have multiple correlated performance characteristics (PCs). The proposed model is asymmetric and capable of capturing a strongly structured physical dependence between the two PCs. Several statistical properties of the model are obtained, and the reliability function and the remaining useful life (RUL) distribution are derived. The penalized likelihood approach is utilized to derive the maximum likelihood estimators for the model parameters and the reliability function. The generalized confidence intervals for these quantities are then constructed using the generalized pivotal quantity method, as well as the generalized prediction intervals for the RUL. In addition, bootstrap-p methods are adopted to construct both confidence intervals and prediction intervals. The performance of the proposed estimation methods is evaluated through Monte Carlo simulations. Finally, the polymeric coating data set is analyzed to demonstrate the applicability and usefulness of the proposed framework.
