Conference Agenda

Knowledge management has become fundamental for organizations, and generative artificial intelligence is radically transforming its concept. This article explores the innovative applications of generative artificial intelligence, highlighting its ability to create original content and optimize traditional processes for identifying, organizing, storing, distributing, and applying organizational knowledge. A generative artificial intelligence-based model is proposed, developed through applied research and validated with cases of Colombian companies already using these technologies. The model integrates the knowledge cycle with elements of the EKMF framework. Challenges such as information validation and ethics in the use of automated models are also addressed. This analysis provides insight into how generative artificial intelligence complements and transforms traditional approaches to knowledge management.

This study presents a predictive system based on Machine Learning and Deep Learning techniques aimed at supporting early prevention of sports injuries in athletes through anthropometric and morphological indicators. The CRISP-DM methodology was adopted to structure the analytical process, incorporating data preprocessing steps such as cleaning, Min–Max normalization, ordinal encoding, class balancing, and stratified train–test splitting. Six predictive models were developed and evaluated: Random Forest, XGBoost, Multilayer Perceptron (MLP), Support Vector Machine, K-Nearest Neighbors, and a Deep Learning model implemented in Keras. In addition, an Ensemble Voting approach was integrated to improve robustness by combining model predictions through majority voting. Model performance was assessed using accuracy, precision, recall, F1-score, and Precision–Recall curves to address class imbalance. The results indicate that the MLP model achieved the highest overall performance, reaching an accuracy of 87.5%, followed by Random Forest and Keras models with competitive results. Finally, the proposed system was deployed as a web-based application that allows sports professionals to input athlete data and obtain real-time injury risk predictions, facilitating preventive decision-making in load management and reducing the likelihood of muscle injuries.

Citrus diseases such as Huanglongbing (HLB), citrus canker, and Citrus Tristeza Virus (CTV) cause substantial economic losses, yet conventional detection methods remain costly and inaccessible to smallholder farmers in developing regions. This paper presents a modular IoT-TinyML architecture for early disease detection that integrates low-cost sensors with edge-based machine learning. The three-layer architecture (Perception, Processing, Application) enables real-time environmental monitoring and on-device visual classification using dual TinyML models for leaf and fruit analysis. A confidence-based decision fusion strategy balances sensitivity and specificity. Solar power and lightweight protocols ensure energy autonomy and remote operability without internet dependency. This edge-computing approach reduces detection latency and costs compared to cloud-based solutions, offering a scalable and practical precision agriculture tool for resource-constrained environments.

The use of artificial intelligence (AI) tools is growing in many areas of human development, particularly in higher education. Among its benefits are personalized learning, adaptation of content and pace to students' individual needs, continuous feedback, improved monitoring of academic progress, and strengthened student autonomy and motivation. However, concerns are also arising among educators and other relevant stakeholders, particularly regarding plagiarism, academic integrity, data privacy, and inequitable access to these technologies. Considering the importance of education for the development of countries, this study reviews the literature on the application of artificial intelligence in higher education in Latin America and the Caribbean, specifically in engineering programs. Analyzing the opportunities and challenges arising from the use of AI is fundamental to guiding the formulation of strategies that promote the responsible, equitable, and contextualized adoption of these technologies, thereby strengthening educational and social development in the region.

In the context of Industry 4.0, Generative Artificial Intelligence (GAI) has primarily been used to automate operational processes and schedule activities. However, the work order (WO) closure stage is often considered a low-value administrative process, despite its potential as a critical source of unstructured knowledge. This article demonstrates the effective use of GenAI in completing maintenance report documentation for strategic industrial assets.

The research proposes a five-stage approach involving case identification, data collection, prompt design, execution using large-language models (LLMs), and technical validation. Two cases reinforces how prompt engineering can extract value from manual records and checklists by transforming fragmented technical descriptions into structured, coherent reports.

The results show that GenAI can be used to leverage closure information for feedback in analytical models and data-driven decision-making. By significantly reducing man-hours and improving the quality of historical data, GenAI acts as a key enabler for transitioning to predictive maintenance. In conclusion, integrating GenAI into the document workflow enables the utilization of previously untapped operational knowledge, thereby strengthening industrial reliability.

In a context where artificial intelligence and large language models (LLMs) are radically transforming education, there is a pressing need to automatically evaluate and organize the content generated by these technologies—particularly assessment questions. This work addresses the technical challenge of classifying LLM-generated questions based on their difficulty, a problem often overlooked by traditional approaches that do not explicitly model the semantic relationship between the question and its source text. We propose a system that generates questions from input passages using LLMs such as Gemini, and classifies them via deep learning models trained with embeddings and regularization techniques, implemented in TensorFlow and PyTorch. Our methodology includes the creation of a custom dataset derived from SQuAD passages, the vectorization of texts and questions using various embedding strategies, and a comparative evaluation of multiple classification architectures. Experimental results show that the model based on paraphrase-MiniLM-L6-v2 achieves 90% bi-class accuracy. This supports the hypothesis that more difficult questions, due to their less ambiguous patterns, are classified with higher precision.

Since beginning design, the designer has used his hands, wits, and pencil to transform an idea into a functional product that serves its purpose. With advances in Deep Learning and convolutional neural networks, a new tool has emerged: generative models of images from text. With the purpose of testing, evaluating and directing these models to apply them to the design of new products, we have focused on a specific product field, olive oil containers, a fairly broad and perfect field for the use of these models, as they are only capable of generating an image; they cannot provide measurement values or volumes of the object. Therefore, oil containers are a perfect starting point: everyday products whose shapes and aesthetics vary by brand and desired shape. The aim is to use these new tools and give them their place of application in the conceptual product design process in the state-of-the-art field of product development. In this work, we will examine the models available on the market and the techniques that can guide them. We will evaluate the results from different image evaluation methods generated by artificial intelligence and the assessments obtained, to propose a methodology and a model for the execution of new conceptual designs.