This study identifies key indicators for strategic biodiversity decisions to improve the effectiveness of environmental ecosystem investments. A novel four-stage model is proposed. First, expert weights are determined using an artificial intelligence-based decision-making method. Second, missing evaluations are estimated for the strategic biodiversity decisions in ecosystem investments using an expert recommender system. Third, criteria weights for strategic biodiversity decisions are computed using quantum picture fuzzy rough set-based modified SWARA (M-SWARA). Finally, investment alternatives are ranked with QPFR-VIKOR. The main contribution of this study is the integration of artificial intelligence into fuzzy decision-making analysis to compute expert weights objectively, thereby enhancing the effectiveness and reliability of the proposed model. Results reveal that technological innovation and financial evaluation are the most significant indicators for strategic biodiversity decisions. Restoration projects and eco-friendly infrastructure are identified as the most suitable investment alternatives. The proposed framework provides managers and policymakers with a transparent, data-driven decision support tool for prioritizing technological capability development and financial feasibility analysis in biodiversity-oriented investment planning, contributing to more sustainable and strategically aligned ecosystem management practices.
Ensuring structural integrity in buildings and infrastructure under extreme loading conditions represents a pivotal challenge in modern civil engineering. Exposure to natural disasters, accidental impacts, and deliberate attacks can result in the application of unprecedented stresses, which may ultimately lead to progressive collapse and catastrophic failures. While traditional analytical methods are reliable, they often prove inadequate in meeting the increasing demand for rapid and accurate assessments in complex scenarios. However, recent advances in computational tools, particularly machine learning (ML), offer a new approach to address these challenges. In this study, nonlinear static analyses of 250 reinforced concrete systems are conducted within pushdown procedures. Load factor and vertical drift capacities of systems are obtained and accepted as target outputs for ML based predictions. By leveraging data-driven models, it becomes possible to predict structural behavior under extreme conditions with greater precision and efficiency. This study builds on this emerging field, aiming to provide novel insights into collapse mechanisms and robustness through advanced machine learning techniques.
Effective inventory management is essential in the food industry due to high demand and the perishable nature of its products. Proper management not only affects daily operations but also plays a crucial role in a company's overall success. This study aims to identify and analyze strategies for improving inventory management by integrating traditional methods with modern technologies. To achieve this, a systematic review of studies published between 2020 and 2024 was conducted using databases such as ScienceDirect and Scopus. 38 relevant articles were selected for analysis based on defined inclusion and exclusion criteria. The findings reveal that adopting advanced technologies, such as automation, artificial intelligence-based management systems, and the Internet of Things (IoT), can greatly improve operational efficiency and reduce inventory losses. Furthermore, the study highlights the importance of strategies like LIFO (Last In, First Out) and FIFO (First In, First Out) for the effective management of perishable items. In conclusion, integrating traditional strategies with technological innovations offers a comprehensive approach to enhancing inventory management within the food sector. This research offers valuable insights for researchers and industry professionals aiming to increase operational efficiency and gain a competitive advantage in the market.
Harmonic distortion is one of the most significant electromagnetic disturbances in power systems due to its widespread propagation and well-known adverse effects, including increased energy losses, accelerated equipment aging, and operational failures. In recent years, this disturbance has intensified due to the increasing penetration of power electronic devices (PEDs) in electrical networks, particularly in the residential sector. Although different models have been developed to analyze the individual harmonic behavior of these devices, the models commonly used to represent aggregated residential loads are often simplified and fail to accurately capture their aggregate harmonic behavior. In this paper, based on experimental measurements, Least Squares Estimation (LSE) is used to develop an individual Coupled Norton (CN) model for a typical set of residential loads, which is then used to construct a component-based aggregated model. The accuracy of this aggregated model is evaluated against a measurement-based aggregated model for the same set of PEDs, showing good performance with errors below 10% in the elements of the Frequency Coupling Matrix (FCM). These results indicate that the proposed model is a viable and accurate alternative for representing aggregated residential loads and constitutes a valuable tool for harmonic analysis studies.
Recent advances in deep learning have improved the automatic detection of cardiac arrhythmia from electrocardiogram (ECG) signals, supporting early diagnosis and treatment. However, most high-performing models require large amounts of labeled data, whose acquisition is costly and dependent on expert annotation. To improve data efficiency in low-label settings, this work proposes a clustering-based strategy for selecting informative unlabeled ECG signals prior to contrastive self-supervised pretraining. The selected signals are used to construct a more representative pre training set, followed by contrastive learning and supervised fine-tuning with a limited number of labeled examples. The proposed approach was evaluated on the Icentia11k dataset, using standard arrhythmia detection metrics, where it consistently outperformed baseline contrastive learning methods. In particular, an average improvement of 4.9 points in the F1-score and 0.4 points in the Area under the Curve (AUC) metric is achieved when only 5000 labeled ECG segments are available. These results demonstrate that targeted unlabeled data selection strengthens contrastive pre-training, leading to improved performance and robustness in data-constrained ECG analysis scenarios.
Fortifying foods with iron is a strategy for combating anemia, but it can alter the sensory and structural properties of extruded products. This study evaluated extruded formulations based on quinoa and corn fortified with heme iron (HI) and analyzed their sensory attributes, gelatinization index (GI), and microstructure, compared to a control sample without iron fortification. Significant differences were observed in aroma (p = 0.0102), color (p < 0.001), and appearance (p < 0.001) between the control sample (S0) and the fortified formulations, with S0 obtaining the highest sensory acceptance. However, texture (p = 0.6763) and flavor (p = 0.8754) did not show significant variations. The GI (94 +/- 0.05) of the control sample was significantly higher (p < 0.001) than that of the fortified samples, indicating a combined effect of fortification and the composition of the cereal and pseudocereal on starch gelatinization. SEM micrographs revealed an amorphous and cohesive matrix, indicative of starch gelatinization during extrusion. Although exploratory, the results suggest that HI fortification can improve the nutritional profile of extrudates, while primarily affecting sensory acceptance in terms of aroma, color, and appearance, highlighting the need to adjust the formulation and extrusion to balance nutritional value and sensory acceptability.
Understanding groundwater recharge patterns is essential for managing water resources in regions facing climate variability and increasing water use pressure. In this study, stable isotopes of Deuterium (2H) and Oxygen-18 (18O) are used as tracers to explore recharge dynamics and surface-groundwater interactions in the Bogota River basin. Using meteorological, geospatial, and isotopic data, the Isocompy library was applied to model the spatial distribution of the isotopic composition of precipitation, resulting in the first local-scale isoscape developed for the basin. The precipitation isoscape revealed a clear spatial variability associated with precipitation levels, recycled moisture, and topographic effects. By comparing this isoscape against the isotopic composition of groundwater, distinct recharge patterns and degrees of aquifer exploitation were identified across the basin. These findings highlight the value of isotopic tools and precipitation isoscapes in identifying recharge zones, understanding hydrological connectivity, and informing sustainable groundwater management. Continuous sampling and improved spatial resolution in isotopic models are recommended to better capture the complexity of recharge processes in heterogeneous basins such as that of the Bogota River.
The use of high-sulfur crude oil in power generation plants has increased SO2 emissions and reduced the service life of steam generators due to corrosive effects in the low-temperature zone, with consequent economic and environmental impacts. The introduction of SO2 reduction technology using CaO requires an understanding of how the reacting system develops, its governing parameters, the stages controlling the reaction rate, and the kinetic model to which it conforms, among other critical aspects. The objective of this study was to determine the kinetic parameters characterizing the reaction between SO2 and CaO at low temperatures, as described by the unreacted core model. In a laboratory-scale facility, the reaction temperature was set at 200, 300, and 400 degrees C, and the SO2 concentration in air at 2%, 4%, and 6%. Gas-film resistance and gas diffusion within the solid bed were eliminated. The results showed that the rate-controlling step is the resistance of the product layer. SO2 reduction increased with rising temperature and SO2 concentration, and CaO conversion did not exceed 20%. The pre-exponential factor was 2.938 & times; 10-14 m2/s, and the activation energy was 17.23 kJ/mol, values characteristic of processes limited by physical diffusion.
Unmanned Aerial Vehicles (UAVs) have become indispensable in fields such as disaster response, precision agriculture, environmental monitoring, and surveillance. Their ability to navigate complex and dynamic environments makes them essential for autonomous operations. However, ensuring accurate and reliable state estimation remains a significant challenge, particularly in GPS-denied environments, where traditional navigation systems suffer from drift, localization errors, and trajectory inconsistencies. Addressing these limitations is crucial for improving UAV autonomy and operational efficiency. This study aims to enhance UAV autonomy and trajectory optimization by integrating a Satellite-Based Augmentation System (SBAS) with monocular visual-inertial odometry (VIO) within a factor graph optimization framework. The proposed methodology fuses visual and inertial sensor data, incorporating state constraints and prior knowledge to improve localization accuracy, reduce drift, and manage uncertainty. Experimental evaluations were conducted under different path optimization conditions to assess system performance. Results show that Path 1 achieved the highest optimization score of 0.481, Path 2 showed moderate optimization at 0.130, while Paths 3, 4, and 5 exhibited minimal improvements, with scores of-6.176, -0.041, and-0.113, respectively. These findings confirm the effectiveness of the proposed approach in optimizing UAV trajectories and enhancing real-time navigation accuracy. The study concludes that integrating SBAS with VIO significantly enhances UAV state estimation, offering a promising solution for autonomous aerial operations in both indoor and outdoor environments. This approach provides a robust and scalable framework for improving UAV navigation in critical applications, ensuring greater reliability under GPS-denied conditions.
The irruption of artificial intelligence (AI) in higher education – especially its generative forms, i.e., models capable of producing text, images, or code – has rapidly transformed pedagogical, administrative, and assessment practices in our institutions. We are facing a reconfiguration of the university space that cannot be reduced to the adoption of tools; instead, it demands a critical examination of its epistemic, ethical, and pedagogical implications. Generative AI has established a set of practices that, if not continuously monitored, could compromise the very essence of education in engineering and other disciplines [1].
This work proposes a technique for placing electric vehicle charging stations using a bootstrapping-based probabilistic power flow. The methodology employs maximum likelihood estimation to model uncertainties in EV charging demand and establish robust confidence intervals for key system metrics. This approach was implemented in the Matpower simulation software within the IEEE-14 bus system, modeling the probabilistic load profile of 4500 EVs while considering 4000 realizations to obtain a wide spectrum of operation scenarios. The main results identified bus 9 as the optimal location for EV charging infrastructure, obtaining minimal active power losses (26.5 +/- 0.5 MW) and a maximum efficiency of 92.87 +/- 0.08 %. The strategic placement of charging stations is closely linked to the lowest active power losses, offering optimal efficiency. However, beyond an optimal placement, this paper aims to increase the robustness of modern grids, overcoming drawbacks related to the integration of electromobility infrastructure. The selection of the most representative features, combined with uncertainty analysis, contributes to an improved decision-making, emphasizing the need for supporting sustainable mobility.
Augmented reality (AR) is an emerging technology that enhances interactive educational modules by helping students to visualize complex concepts, thereby improving engagement and academic performance. This study aimed to assess the effects of an AR-based educational module on students’ cognitive engagement during online learning, specifically focusing on liver cancer cell characterization as an extension of the cell structure topics in the biology syllabus. A total of 104 students were divided into three groups: one using the AR module, another using a conventional module, and a control group. The students’ performances were assessed through pre-test and post-test analyses. Additionally, the facial features of the AR group were analyzed using MediaPipe’s Face Mesh algorithm to calculate the eye aspect ratio (EAR) which determines attention levels based on the opening and closure of the eyes. The results showed that the AR-based module had a positive effect on students' performance (9.73%) compared to the conventional module and effectively captured students' attention during the first half of the lesson. This integration of technology is beneficial for enhancing online teaching practices, particularly when it comes to engaging biomedical engineering students with challenging biology topics like cancer cell characterization.
This article reports the results of a study that examined the wear and thermal behavior of Al/ZrO2 nanocomposites fabricated using powder metallurgy with different ZrO2 reinforcement contents (3, 6, and 9 wt.%). The findings show that microhardness increased to 33 HV in sample AZ0, 58 HV in AZ3, 74 HV in AZ6, and 87 HV in AZ9, indicating that higher ZrO2 content leads to enhanced surface resistance. The tensile strength of sample AZ0 was 285 MPa, with elongation values decreasing to 6%, which demonstrates a transition from more ductile performance to increased mechanical strength with the addition of reinforcement material. Compressive strength also exhibited a significant improvement, increasing from 43 MPa in AZO to 376 MPa in AZ9, indicating enhanced load-bearing capacity in the reinforced composite. The coefficient of friction decreased markedly from 1.6 to 0.4, reflecting improved wear resistance due to the homogeneous distribution of ZrO2 nanoparticles and the formation of a hard ceramic phase. Thermal conductivity also decreased from 237 Wm-1K-1 to 150 Wm-1K-1, which is attributed to the low thermal conductivity of ZrO2 and its homogeneous incorporation into the matrix. Similarly, the coefficient of linear thermal expansion decreased from 22x10-6/K to 11x10-6/K, owing to the thermal barrier effect and dimensional stability provided by the ceramic reinforcement. Overall, these results demonstrate that ZrO2 nanoparticles have the potential to enhance the mechanical strength, wear resistance, and thermal stability of aluminum nanocomposites.
In the reviewed literature, experimental and simulation methods prevail as the main tools for developing technologies for converting plastic waste into synthetic fuels. This paper proposes a methodology for determining the operating parameters for this conversion using high-density polyethylene (HDPE) as an example. To address the problem, a methodological approach based on engineering systems analysis and synthesis was used. This resulted in a conceptual mathematical model for developing HDPE pyrolysis technology, both general for process operation and specific for a basic facility design. A virtual experimental plan was executed using the Aspen Plus process simulation tool under thermodynamic equilibrium conditions. Based on the generated data, approximation functions for the efficiency indicators and the constrained functions required for parameterization were developed. Decisions were then made to determine the optimal pyrolysis temperature by iteratively optimizing the resulting detailed nonlinear multi-objective model, varying the desired values of the efficiency indicators. This procedure allows for the identification of a pyrolysis temperature that satisfies the preferences of potential decision-makers and establishes the conditions for addressing the general problem posed for the operation of the conversion process.
Precipitation estimation at the river basin level is essential for watershed management, the analysis of extreme events and weather and climate dynamics, and hydrologic modeling. In recent years, new approaches and tools such as artificial intelligence techniques have been used for precipitation estimation, offering advantages over traditional methods. Two major paradigms are artificial neural networks and fuzzy logic systems, which can be used in a wide variety of configurations, including hybrid and modular models. This work presents a literature review on hybrid metaheuristic and artificial intelligence models based on signal processes, focusing on the applications of these techniques in precipitation analysis and estimation. The selection and comparison criteria used were the model type, the input and output variables, the performance metrics, and the fields of application. An increase in the number of this type of studies was identified, mainly in applications involving neural network models, which tend to get more sophisticated according to the availability and quality of training data. On the other hand, fuzzy logic models tend to hybridize with neural models. There are still challenges related to prediction performance and spatial and temporal resolution at the basin and micro-basin levels, but, overall, these paradigms are very promising for precipitation analysis.
In the literature, experimental data on X-shaped screw connections have been analyzed in order to develop an empirical model for their shear force capacity and stiffness, which are important parameters in designing timber-concrete composite structures. Although considerable research has been conducted worldwide to understand the composite action of timber and concrete, there is no generic model for determining the shear force capacity and stiffness of screw connections; most of the existing models are based on theoretical derivations. In this paper, empirical models are derived to determine the shear capacity of screw connections installed in X-shaped arrangements, considering the embedment and withdrawal strength of the screws within the timber and concrete. Moreover, a stiffness model based on global flexibility, as influenced by the material properties of timber, concrete, and screws, is elaborated. The model is validated using existing push-pull data and variations in material properties. A comparison with a well-known model demonstrates the suitability of our proposal. This model can be used to predict the shear force capacity and stiffness of X-shaped screw connections in timber-concrete composite structures.
The aim of this study was to investigate how productivity indicators in the Colombian dairy industry can be strengthened with quality management practices (QMp). This research was conducted under a multiple case study approach, analyzing two companies within the Colombian dairy industry. The primary data collection method used was in-depth interviews, supported by secondary sources such as internal documents and company websites. According to the main findings, the interviewees perceive that the key QMp to strengthen the seven productivity indicators of the Colombian dairy industry include all practices related to top management support, human resources management, process management, and process control, as well as individual practices like innovation and feedback and auditing. Furthermore, the indicators with the highest number of key QMp were those related to the amount of whey and defective product. This research provides valuable and original insights for practitioners, managers, and policymakers. In addition, it contributes to theoretical progress by outlining how specific productivity indicators within the Colombian dairy industry can be improved through key QMp, proposes a theoretical framework that suggests the possible sequence of implementation and the interrelationships between these practices, and lays a foundation for future research in this field.
This editorial reflects on the incorporation of artificial intelligence into scientific publishing based on the experience of the journal Ingeniería e Investigación. It examines the main tensions arising from the use of AI tools across editorial workflows, including desk review, peer review, language editing, visual production, and metadata and format management. The editorial argues that AI should be understood as a technical support tool under human supervision, aimed at improving efficiency without replacing critical judgment or editorial responsibility. It concludes by emphasizing the need for clear policies, transparency in AI use, and editorial literacy processes to ensure its integration without compromising scientific integrity.
This study uses an explanatory mixed-methods design to develop and validate a DT-PM (digital twins-project management) maturity framework. To this effect, it combines a cross-sectional survey of 200 professionals working in PM, six in-depth case studies, and, as part of a design science research (DSR) cycle, a representative sample of individuals working in PM. The results extend the technology-organization-environment (TOE) framework by incorporating time-based project factors such as stakeholder movement and workflow plasticity. Taken together, these factors account for 71% of the difference in implementation success. The analysis reveals significant sociotechnical contradictions with direct effects on PM practice, namely an authority paradox and a 15% threshold phenomenon for initial project viability, which provides managers with a clear way to assess whether benefits are being realized. The maturity framework validated in this study is an organized diagnostic tool that ensures that DT capabilities are aligned with project lifecycle stages and PM knowledge areas. This study concludes that successful DT adoption in the cases examined entails not only upgrading technology but also addressing sociotechnical alignment. This involves moving from a technology-centered implementation towards adaptive project governance and organizational learning in environments with limited resources.
Biorefineries have emerged as crucial elements in the circular economy, offering a sustainable solution for converting residual biomass into diverse valuable bioproducts by integrating various biotechnological pathways. Despite the challenges posed by the scale of animal waste production, biorefineries have demonstrated their ability to overcome these obstacles and unlock the inherent value of these resources. In this work, a comprehensive bibliometric analysis of specialized literature and patents on the valorization of pig manure was conducted. Among the various techniques, anaerobic digestion (AD) emerged as the most promising method for waste valorization, serving as a platform for biorefinery conceptualization. AD enables the segregation of biorefinery streams and exhibits considerable potential for generating a wide array of subproducts. The relationship between production and environmental indices has been established worldwide. This work proposes a conceptual biorefinery model that incorporates relevant biotechnological routes for the identified bioproducts. These include biogas, hydrogen, electricity, microalgae, bioethanol, volatile fatty acids, organic amendments, biofertilizers, and biodiesel. The limitations and advantages of the most significant processes have been duly considered and included in the model.
