Cristian García-Pumagualle

The supervision of inclusive financial institutions in emerging economies represents a critical challenge for financial stability. While cooperatives and mutualists are pillars of financial inclusion, there is a lack of explainable early warning models adapted to their structural risk profiles. This study evaluates an explainable machine learning framework for early supervision, utilizing a longitudinal dataset covering 584 unique financial institutions (cooperatives and mutualists) with 24,232 monthly observations across the 2018–2024 predictor observation window (yielding 17,799 healthy transition candidate observations and 74 evaluated distress episodes: 68 in the training set and 6 in the out-of-time validation set). Adhering to ISO/IEC 25012 data quality standards and a six-month forecast horizon (T + 6), the performance of an XGBoost-based model was evaluated against traditional logistic regression benchmarks. Given the extreme class imbalance (0.1556% event rate in the out-of-time test set), model utility was assessed primarily through the precision–recall AUC (PR-AUC). The results demonstrate that XGBoost achieved a higher PR-AUC point estimate (0.2466) than the logistic regression benchmark (0.0931) in the 2024 out-of-time validation sample. The use of SHAP explainability revealed distinct risk drivers for cooperatives and mutualists, providing exploratory support for segmented supervisory monitoring. This study proposes an interpretable decision support framework that enables proactive, risk-based prioritization in a dollarized emerging economy.

This study introduces AgroTec 4.0, which is a smart farming system designed to revolutionize strawberry cultivation in greenhouses through the integration of edge computing technology in the Andean region of Ecuador. The primary objective has been to enhance cultivation efficiency by comparing results from strawberry crops with and without the system, under identical greenhouse conditions. Given the low educational and economic status of local farmers, AgroTec 4.0 was engineered to be user-friendly, easy to operate, and cost-effective, empowering producers with data-driven decision-making capabilities. Key findings underscore the potential of AgroTec 4.0 and agricultural data, including a 15% increase in strawberry yield, from 5.0 kg/m² in the control greenhouse to 5.75 kg/m² with AgroTec 4.0, highlighting the system’s ability to maximize productivity. There has also been a significant 20% reduction in water usage, decreasing from 80 L/m² in the control greenhouse to 64 L/m² with the system, showcasing AgroTec 4.0’s efficiency in resource management. Furthermore, there were significant improvements in fruit quality, with an 11.8% increase in the Brix index (from 8.5 to 9.5) and a 16.7% increase in average fruit weight (from 30 to 35 g), demonstrating the system’s capacity to enhance product quality. Finally, there has been an impressive 103.03% return on investment (ROI) with AgroTec 4.0, compared to no change in ROI in the control greenhouse, emphasizing the economic value of implementing this technology. These results underscore the transformative potential of AgroTec 4.0 in precision agriculture, offering a scalable and sustainable approach for small-scale producers in Ecuador. The system’s modularity and real-time data analysis capabilities allow for flexible adaptation to various needs, providing farmers with an intuitive interface for managing crops and optimizing resource use. This study demonstrates the feasibility of leveraging agricultural data and edge computing to improve cultivation practices and enhance productivity, contributing efficiently to the sustainability of agriculture in challenging environments.

The current study aims to analyze the fundamental ecological niches of five páramo species in the high Andean zones of South America. Hereby, the research focuses on the typical flora species of this ecosystem, such as Azorella aretioides, Azorella crenata, Azorella biloba, Azorella corymbosa, and Azorella pedunculata, testing whether their niches are more similar than expected. The study area in the Ecuadorian Andes encompasses altitudinal gradients between 2000 and 5000 m above sea level, where the species were previously registered. For the methodology, we used data from the Global Biodiversity Information Facility (GBIF) and identified the determining bioclimatic variables for the model as Bio1, Bio2, Bio3, Bio4, Bio12, Bio13, Bio14, Bio18, and Bio19. Moreover, we considered two alternative socioeconomic scenarios of climate change (SSP126 and SSP 245) projected for the period 2021–2040. In the same way, ecological niche models were evaluated and applied in order to identify areas of reduction in geographic space and the ecological space of the five species. Subsequently, the analysis revealed significant sensitivity of these high mountain species to variations in temperature and precipitation, with potential risks to associated species and water resources. In conclusion, changes in the fundamental niche were observed, with all species indicating reductions in distribution for 2040. This occurs particularly under the SSP245 scenario and especially in the equatorial zone, which currently exhibits extraordinary richness concerning the Azorella genus. Therefore, the study highlights evidently the vulnerability of páramo species to climate change, emphasizing the need for conservation efforts.

Este estudio presenta una solución sostenible para optimizar el uso del agua en la producción ganadera. Los resultados presentados muestran la eficiencia en el uso del agua y la mejora en el rendimiento de los pastos mediante la implementación de un sistema de riego automatizado en una hacienda ganadera en Ecuador. Se encontró que la implementación del sistema permitió una reducción del 18,80% en el consumo promedio diario de agua, lo que representa una significativa optimización del recurso hídrico. Además, se observó una mejora en la uniformidad de riego y en la capacidad de carga forrajera de los pastos. El sistema de riego automatizado permitió la recolección de datos de humedad en tiempo real y la visualización de estos de manera remota a través de una aplicación web. Además, los costos de implementación del sistema resultaron ser un 2,16% más económicos respecto al promedio de los sistemas comerciales convencionales, lo que demuestra su rentabilidad. En conclusión, este estudio demuestra que la implementación de un sistema de riego automatizado puede mejorar significativamente la eficiencia en el uso del agua y la capacidad de carga forrajera de los pastos en una hacienda ganadera. Además, se destaca que este tipo de sistemas son rentables y pueden ser una solución sostenible para optimizar el uso del recurso hídrico en la producción ganadera.