In order to scientifically evaluate the germplasm resources of Momordica charantia in southern China, the diversity, correlation and cluster analysis were carried out on the main botanical characters of 56 Momordica charantia varieties, such as melon length, melon transverse diameter, single melon weight, internode length, stem diameter, leaf length and leaf width. The results showed that the variation coefficients of 7 agronomic characters of 56 Momordica charantia varieties ranged from 8.81% to 19.44%, the average variation coefficient was 14.21%, the maximum variation coefficient of single melon weight was 19.44%, and the minimum variation coefficient of melon cross diameter was 8.81%. The correlation analysis showed that there were correlations among the agronomic traits. The positive correlation coefficient between leaf length and leaf width was up to 0.978, and the negative correlation coefficient between single melon weight and internode length was up to 0.451. The 56 varieties were divided into 3 groups by cluster analysis, of which 92.86% of the materials were concentrated in the first and second groups, and there were only 4 materials in the third group. The results can provide a reference for the cultivation, utilization and genetic improvement of Momordica charantia resources in southern China.

This research introduces a novel framework integrating stochastic finite element analysis (FEA) with advanced circular statistical methods to optimize heat pump efficiency under material uncertainties. The proposed methodologies and optimization focus on balancing the mean efficiency and variability by adjusting the concentration parameter of the Von Mises distribution, which models directional variability in thermal conductivity. The study highlights the superiority of the Von Mises distribution in achieving more consistent and efficient thermal performance compared to the uniform distribution. We also conducted a sensitivity analysis of the parameters for further insights. The results show that optimal tuning of the concentration parameter can significantly reduce efficiency variability while maintaining a mean efficiency above the desired threshold. This demonstrates the importance of considering both stochastic effects and directional consistency in thermal systems, providing robust and reliable design strategies.

The purpose of this study is to predict the frequency of mortality from urban traffic injuries for the most vulnerable road users before, during and after the confinement caused by COVID-19 in Santiago de Cali, Colombia. Descriptive statistical methods were applied to the frequency of traffic crash frequency to identify vulnerable road users. Spatial georeferencing was carried out to analyze the distribution of road crashes in the three moments, before, during, and after confinement, subsequently, the behavior of the most vulnerable road users at those three moments was predicted within the framework of the probabilistic random walk. The statistical results showed that the most vulnerable road user was the cyclist, followed by motorcyclist, motorcycle passenger, and pedestrian. Spatial georeferencing between the years 2019 and 2020 showed a change in the behavior of the crash density, while in 2021 a trend like the distribution of 2019 was observed. The predictions of the daily crash frequencies of these road users in the three moments were very close to the reported crash frequency. The predictions were strengthened by considering a descriptive analysis of a range of values that may indicate the possibility of underreporting in cases registered in the city’s official agency. These results provide new elements for policy makers to develop and implement preventive measures, allocate emergency resources, analyze the establishment of policies, plans and strategies aimed at the prevention and control of crashes due to traffic injuries in the face of extraordinary situations such as the COVID-19 pandemic or other similar events.