Cristian Darwin Borja

Prof. Cristian Darwin Borja

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications

20 June 2024 in Salud, Ciencia y Tecnología
Verified
Journal Article
Open access
PDF file
ENElian Jair Ordoñez Ninasunta
Elian Jair Ordoñez Ninasunta
DRDarwin Yazmany Paez Ruiz
Darwin Yazmany Paez Ruiz
Cristian Darwin BorjaCristian Darwin Borja
Cristian Darwin Borja
JZJOHNNY BAJAÑA ZAJIA
JOHNNY BAJAÑA ZAJIA
The research project focuses on developing an offline mobile application to explore the tourist routes of Malqui Machay, offering information about access roads and tourist establishments in the area. The Mobile-D methodology was used, interviews and surveys were carried out to collect information about tourists and technological needs. The lack of internet coverage in the area motivated the creation of an offline tool. The project, in collaboration with the Technical University of Cotopaxi, seeks to improve access to Malqui Machay and enhance its tourist attraction. Quantitative research, literature review, case study, field and technological research supported the development of the application, offering a solid theoretical basis and tools for its creation. Staying up to date with technological trends is crucial for future improvements in mobile applications

5 August 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
William Paul Pazuña NaranjoWilliam Paul Pazuña Naranjo
William Paul Pazuña Naranjo
MRMilton Andrés Bautista Romero
Milton Andrés Bautista Romero
FBFreddy Rodrigo Romero Bedón
Freddy Rodrigo Romero Bedón
WHWilliam Armando Hidalgo
William Armando Hidalgo
PVPaco Jovanni Vásquez-Carrera
Paco Jovanni Vásquez-Carrera
DGDario Xavier Cobos Guayaquil
Dario Xavier Cobos Guayaquil
This study presents a spatiotemporal analysis and predictive modeling approach for assessing distributed photovoltaic potential in La Maná, Ecuador, using hourly NASA POWER satellite data from 2014 to 2025. A total of 105,168 hourly records of global horizontal irradiance, air temperature, relative humidity, and wind speed were processed. The methodology included data cleaning, temporal feature generation, hourly, monthly, and interannual solar resource analysis, and the comparison of machine learning models for predicting global horizontal irradiance as an indicator of photovoltaic potential. The results showed a relatively stable solar resource throughout the study period, with hourly maximum values occurring around midday and low season-al variability. Among the evaluated models, Gradient Boosting achieved the best predictive performance, with an MAE of 35.14 W/m², an RMSE of 68.44 W/m², and a coefficient of determination of 0.8875. The feature importance analysis revealed that the hourly component dominated solar resource prediction; however, when this effect was excluded, air temperature emerged as the most influential meteorological variable. These findings demonstrate that integrating hourly satellite data with machine learning provides a robust alternative for evaluating photovoltaic potential in tropical regions with limited ground-based meteorological monitoring infrastructure.

8 October 2026 in Fluids
Verified
Journal Article
Open access
PDF file
HCHéctor Calvopiña
Héctor Calvopiña
JEJosé Erazo
José Erazo
Cristian Darwin BorjaCristian Darwin Borja
Cristian Darwin Borja
LRLuis Ramirez
Luis Ramirez
JEJavier Aldair Estevez
Javier Aldair Estevez
DJDavid Eduardo Balseca Jaramillo
David Eduardo Balseca Jaramillo
Cyclone separators are widely used for gas–solid separation; however, improving separation efficiency generally increases pressure drop, making cyclone design a challenging multi-objective optimization problem. This study aims to optimize a high-efficiency Stairmand cyclone separator by simultaneously maximizing separation efficiency and minimizing pressure drop through a CFD-based optimization framework integrating Computational Fluid Dynamics (CFD), Design of Experiments (DOE), Response Surface Methodology (RSM), and the Non-dominated Sorting Genetic Algorithm II (NSGA-II). The optimization was performed using two geometric design variables: the vortex finder length (Lv) and the spigot diameter (Du). CFD simulations generated the data required to construct surrogate models, enabling rapid evaluation of candidate geometries during the optimization process. The Pareto-optimal solutions obtained with NSGA-II were subsequently verified using independent CFD simulations before manufacturing the selected configuration as a physical prototype. The optimized cyclone achieved a separation efficiency of approximately 93% while maintaining a low pressure drop of about 60 Pa. Experimental testing showed excellent agreement with the numerical predictions, with a deviation of only 1.51% in separation efficiency, confirming the reliability of the proposed optimization framework. The complete validation chain, from CFD simulation and surrogate-assisted optimization to prototype fabrication and laboratory testing, demonstrates that the proposed methodology provides a practical and reliable approach for the multi-objective design and optimization of industrial cyclone separators.

1 July 2025 in Revista Ingenio global
Verified
Journal Article
Open access
PDF file
HVHenry Stalyn Soledispa Vera
Henry Stalyn Soledispa Vera
ACAngel Julián Navarrete Cedeño
Angel Julián Navarrete Cedeño
Cristian Darwin BorjaCristian Darwin Borja
Cristian Darwin Borja
RMRicardo Augusto Luna Murillo
Ricardo Augusto Luna Murillo
Este artículo presenta el diseño, la capacitación y la validación de una aplicación móvil inteligente para la predicción automatizada del crecimiento y el rendimiento en plantas leguminosas, utilizando algoritmos de aprendizaje supervisado. Se emplearon modelos de bosque aleatorio y árbol de decisión, entrenados con un conjunto de datos multivariante de 319 registros y 67 variables cuantitativas y cualitativas recopiladas a través de sensores IoT y APIs meteorológicas. El modelo de árbol de decisión logró un coeficiente de determinación de 0,76 para la altura de la planta y de 0,87 para el peso del forraje, superando en precisión al modelo de bosque aleatorio, con valores de R² de 0,80 y 0,72, respectivamente. La aplicación, desarrollada en React Native y conectada a un backend de Django, permite al usuario seleccionar el algoritmo que desea utilizar. Además, la validación funcional, realizada con 350 beneficiarios, incluyendo agricultores y estudiantes, mostró un alto nivel de aceptación: el 91% calificó positivamente la usabilidad de la aplicación, y el 84% expresó su intención de usarla recurrentemente. Esta propuesta representa una contribución significativa al ecosistema de la agricultura digital, proporcionando una herramienta accesible, precisa y adaptable con potencial de escalabilidad y adopción por parte de la comunidad.

17 June 2025 in Preprints.org
Verified
Preprint
Open access
PDF file
AJAlex Joel Torres Jimenez
Alex Joel Torres Jimenez
DADennis Brishith Chuqui Alcivar
Dennis Brishith Chuqui Alcivar
Cristian Darwin BorjaCristian Darwin Borja
Cristian Darwin Borja
JZJOHNNY BAJAÑA ZAJIA
JOHNNY BAJAÑA ZAJIA
Cacao production, a key economic pillar for numerous rural communities in Ecuador, faces structural challenges related to climate variability and limited digital connectivity. This study presents the development and implementation of a yield prediction model based on the XGBoost algorithm, integrated into an offline mobile application designed to operate in agricultural environments without internet access. The research followed the CRISP-DM methodology and included the analysis of 5584 observations collected from plots in La Maná (Cotopaxi), corresponding to three cacao genotypes. Variables were processed using cleaning, imputation, and normalization techniques. The predictive model, validated with standard metrics (MSE, RMSE) and an R² of 0.9399, demonstrated robust fit and high interpretability. Subsequently, the model was deployed in a mobile app developed with React Native. Field deployment showed response times under five seconds, compatibility with low-end devices, and high user acceptance. Participatory validation confirmed the practical usefulness of the tool for real-time agronomic decision-making. This work provides evidence of the value of AI tailored to rural contexts and proposes a replicable approach for other value chains under similar conditions.