Civil and Structural Engineering, Artificial Intelligence, Building and Construction, Modeling and Simulation
13
Scopus Publications
Scopus Publications
Estimation of diurnal greenhouse gas (GHG) emissions from unfertilized coffee soils using recurrent neural networks (RNN). A case study for Chirinos, San Ignacio Province, Cajamarca, Peru Wendy Laurent Díaz Saavedra, Jorge Antonio Fernandez Jibaja, Jhon Franklin Oblitas Troyes, Jose Manuel Palomino Ojeda, Manuel Emilio Milla Pino, Annick Estefany Huaccha Castillo, Ruben Eusebio Acosta Jacinto, Guillermo Guardia Vázquez Clean Energy Science and Technology, 2025 Global warming, driven by rising greenhouse gas (GHG) concentrations, has agriculture as a major source of emissions. In coffee plantations, low sampling frequency and the absence of diurnal baselines introduce bias in emission estimates. The objective of this research was to estimate diurnal CO₂, N₂O, and CH₄ emissions from unfertilized coffee soils using recurrent neural networks (RNN). Gas fluxes were measured with a closed dynamic chamber (CDC) at 20-minute intervals between 8:00 and 18:00 over 22 days. For the estimation of GHG emissions, climatic data measured through a meteorological station were used, in addition to environmental parameters incorporated in the CDC. Five RNN models composed of two hidden layers of 20, 25, and 50 neurons were developed, trained, and validated for each GHG. Results indicate that N₂O contributed most to total emissions (734,689 ppm CO₂-eq), with CO₂ (237,579 ppm CO₂-eq) and CH₄ (215,426 ppm CO₂-eq) contributing less. Model performance was strong, with R² values of 0.98 (CO₂), 0.96 (N₂O), and 0.94 (CH₄). It is concluded that the RNNs proved to be reliable models for predicting GHG emissions in unfertilized coffee soils, with this study presenting a replicable framework with the potential to improve temporal estimation and reduce uncertainty in GHG inventories.
Sigmoidal Mathematical Models in the Planning and Control of Rigid Pavement Works Jose Manuel Palomino Ojeda, Lenin Quiñones Huatangari, Billy Alexis Cayatopa Calderon, Manuel Emilio Milla Pino, José Luis Piedra Tineo, Marco Antonio Martínez Serrano, Rosario Yaqueliny Llauce Santamaria Applied Sciences Switzerland, 2025 The objective of the research was to use sigmoidal mathematical models for the planning and control of rigid pavement works. A dataset was constructed using 140 technical files, which were then analyzed to extract the valued work schedules. These schedules contained the variables time and cost per month. Subsequently, two groups were created from the dataset: a training group comprising 80% of the data and a test group comprising the remaining 20%. Subsequently, the variables were normalized and adjusted with the proposed logistic, Von Bertalanffy, and Gompertz models using Python 3.11.13. Following the implementation of training and validation procedures, the logistic model was identified as the optimal fit, as indicated by the following metrics: R2 = 0.9848, MSE = 0.0026, RMSE = 0.0506, and MAE = 0.0278. The implementation of the aforementioned model facilitates the establishment of an early warning system with a high degree of effectiveness. This system enables the evaluation of the discrepancy between the actual progress and the planned progress with an R2 greater than 98%, thereby serving as a robust instrument for the adjustment and revalidation of activities before and following their execution.
Climatic Aggressiveness and Precipitation Concentration in a Peruvian Amazon Basin: Alto Huallaga Interbasin Guillermo Arriola Carrasco, Luis Villegas, Noe Marín, Cesar Idrogo, José Piedra, Jose Palomino, Billy Cayatopa Revista Politecnica, 2025 Precipitation in the central Peruvian Amazon is characterized by being seasonal and with strong intensities during the first months of the year, leading to flooding and the subsequent collapse of the local infrastructure in provinces of San Martin, Huanuco, Pasco and La Libertad Regions of Peru, which are located within the delimitation of the Alto Huallaga Interbasin. Therefore, the objective of this research was to evaluate the climatic aggressiveness and concentration precipitation in the Amazon basin, applying three indices of aggressiveness and a precipitation concentration index (ICP), estimated from the precipitation record of the climatic stations in the study area. The results show a very high correlation between mean precipitation and altitude (R2 = 0.72) and with respect to the aggressiveness, the modified Fournier-Maule index (IFMM) was the one with the best correspondence with respect to altitude (R2 = 0.72) and mean precipitation (R2 = 0.98), however, the ICP shows moderate correlations with altitude (R2 = 0.21) and mean precipitation (R2 = 0.16). Likewise, the extreme values of the different indices were estimated for different return periods and a multiple linear regression model was developed to relate climatic aggressiveness and the ICP to estimate the mean precipitation (R2 = 0.99). Finally, it is concluded that, the Alto Huallaga Interbasin presents a very low climatic aggressiveness and the concentration of rain is moderately seasonal.
Coffee Rust Severity Analysis in Agroforestry Systems Using Deep Learning in Peruvian Tropical Ecosystems Candy Ocaña-Zuñiga, Lenin Quiñones-Huatangari, Elgar Barboza, Naili Cieza Peña, Sherson Herrera Zamora, Jose Manuel Palomino Ojeda Agriculture Switzerland, 2025 Agroforestry systems can influence the occurrence and abundance of pests and diseases because integrating crops with trees or other vegetation can create diverse microclimates that may either enhance or inhibit their development. This study analyzes the severity of coffee rust in two agroforestry systems in the provinces of Jaén and San Ignacio in the department of Cajamarca (Peru). This research used a quantitative descriptive approach, and 319 photographs were collected with a professional camera during field trips. The photographs were segmented, classified and analyzed using the deep learning MobileNet and VGG16 transfer learning models with two methods for measuring rust severity from SENASA Peru and SENASICA Mexico. The results reported that grade 1 is the most prevalent rust severity according to the SENASA methodology (1 to 5% of the leaf affected) and SENASICA Mexico (0 to 2% of the leaf affected). Moreover, the proposed MobileNet model presented the best classification accuracy rate of 94% over 50 epochs. This research demonstrates the capacity of machine learning algorithms in disease diagnosis, which could be an alternative to help experts quantify the severity of coffee rust in coffee trees and broadens the field of research for future low-cost computational tools for disease recognition and classification
Assembly Algorithms for Seismic Vulnerability Estimation in Confined Masonry Dwellings Jose Manuel Palomino Ojeda, Lenin Quiñones Huatangari, Jeiden Revilla Arce, Nilthon Arce Fernández, Marcos Antonio Gonzales Santisteban, Marco Antonio Martínez Serrano International Journal of Safety and Security Engineering, 2024 In Peru, confined masonry houses are self-built, which makes it crucial to determine their seismic vulnerability.The objective of the research was to estimate the seismic vulnerability of confined masonry dwellings in the Pueblo Libre-Ja n sector using assembly algorithms.A database was constructed with data obtained from the National Institute of Civil Defense (INDECI), scientific articles, and theses.Subsequently, the data set was divided into a training set (80%) and a validation set (20%), employing the stacking method with five combinations CB_1, CB_2, CB_3, CB_4, and CB_5.The basic algorithms Gradient-Boosting, Random-Forest, Extra-Tree, and Decision-Tree were utilized as the base algorithms, with the final estimator being the Random Forest Meta-Learner.The models were trained and validated in Python, achieving accuracies of 94. 95, 95.48, 95.39, and 95.66 for the base models and 95.62, 95.23, 95.76, 95.90, and 94.80% for the ensemble models.The most accurate models were the simple Gradient Boosting (95.66%) and the assembled models CB_3 (95.76%) and CB_4 (95.90%).The CB_4 model, which is composed of the Decision Tree and Gradient Boosting algorithms, was applied to the Pueblo Libre sector and yielded a reliability estimate of greater than 95% for the seismic vulnerability of confined masonry.This estimate was classified as high (1.48%),moderate (32.85%), and low (65.67%).It is anticipated that the model implemented will enable engineers and authorities to implement mitigation measures to reinforce housing in the event of a seismic event.
Fuzzy Optimization Model for Decision-Making in Single Machine Construction Project Planning Nilthon Arce Fernández, Flabio Gutiérrez Segura, Manuel Emilio Milla Pino, Jose Manuel Palomino Ojeda, Alfredo Lázaro Ludeña Gutiérrez, River Chávez Santos Mathematics, 2024 Scheduling for a construction project with a limited number of machines is a critical and well-studied problem. Most studies assume that task processing times are exact; in practice, delays frequently occur, rendering the initial work plan invalid. Therefore, adaptability is crucial to the success of a project. This work introduces a fuzzy optimization model for the planning of construction projects executed simultaneously and having only one backhoe. The model assumes imprecise task processing times, represented by triangular fuzzy sets, that accept delays up to a permitted degree of tolerance. The model solution obtains a fuzzy work plan. This is a robust plan that supports incidents (delays). A method to apply the model was created. The fuzzy model can help construction companies reduce delays in the delivery of their projects and avoid excessive penalties. The model was implemented in the CPLEX solver, which can quickly obtain an optimal solution for small and medium instances. For large instances, the model must be solved with metaheuristics. This scientific contribution is important for future work since it consists of the application of fuzzy optimization in a specific area of civil engineering.
Estimation of the Physical Progress of Work Using UAV and BIM in Construction Projects Jose Manuel Palomino Ojeda, Lenin Quiñones Huatangari, Billy Alexis Cayatopa Calderon, José Luis Piedra Tineo, Christiaan Zayed Apaza Panca, Manuel Emilio Milla Pino Civil Engineering Journal Iran, 2024 The delay in the physical progress of construction creates additional costs, missed deadlines, and quality issues. The research aimed to estimate the physical progress of the project by using unmanned aerial vehicles (UAVs) and building information modeling (BIM). The methodology comprised capturing 848 high-resolution images of the Civil Engineering Laboratory construction site at the National University of Jaen, Cajamarca, Peru, using the Phantom 4 RTK drone. The photographs were processed using Agisoft 2.0.1 software, resulting in a point cloud. This was then imported into ReCap Pro 2023 software, which was used to assess the quality of the points. The Revit 2023 software was subsequently utilized to establish the phase parameters, linking the BIM model with the point cloud, filtering the model, and eventually exporting it to the Power BI 2023 software. The work's estimated progress utilizing the proposed methodology was 42.82%, which was not statistically significant compared to the Public Works Information System (INFOBRAS) of 43.14%. This allows for the automation of customary processes, the identification of crucial issues, and prompt decision-making. The study's originality lies in the suggestion of integrating aerial imagery with drones and BIM modeling for the real-time and precise estimation of work progression. This method provides a precise and effective substitute for traditional techniques for gauging the tangible advancement of projects. Doi: 10.28991/CEJ-2024-010-02-02 Full Text: PDF
Convolutional Neural Network for Predicting Failure Type in Concrete Cylinders During Compression Testing Jose Manuel Palomino Ojeda, Billy Alexis Cayatopa-Calderón, Lenin Quiñones Huatangari, Jose Luís Piedra Tineo, Manuel Emilio Milla Pino, Wilmer Rojas Pintado Civil Engineering Journal Iran, 2023 Cracks in concrete cause structural damage, and it is important to identify and classify them. The objective of the research was to describe the behavior and predict the type of failure in concrete cylinders using convolutional neural networks. The methodology consisted of creating a database of 2650 images of failure types in concrete cylinders tested in compression at the Laboratory of Testing and Strength of Materials of the National University of Jaen, Cajamarca, Peru. To identify cracks on the concrete surface, the database was divided into training (60%), validation (20%), and testing (20%), and a transfer learning approach was developed using the MobileNet, DenseNet121, ResNet50, and VGG16 algorithms from the Keras library, programmed in Python. To validate the performance of each model, the following indicators were used: recall, precision, and F1 score. The results show that the models studied correctly classified the type of failure in concrete with accuracies of 96, 91, 86, and 90%, with the MobileNet algorithm being the best predictor with 96%. The novelty of the study was the development of deep learning algorithms with different architectures that can be used in structural health assessment as an automated and reliable method compared to traditional ones. In addition, these trained algorithms can be used as source code in drones for structural monitoring. Doi: 10.28991/CEJ-2023-09-09-01 Full Text: PDF
