vol. 5 núm. 1 (2024): la ciencias de la computación y telecomunicaciones aplicadas con herramientas inteligencia artificial

In the late 1950s, the term “Artificial Intelligence” (AI) was formally introduced at Dartmouth University, where the first step in a new topic of study of how machines simulate human intelligence activities was made. Then, in 2016, the computer “AlphaGo” won a game against the world chess champion; thus, the worldwide interest in artificial intelligence increased. Many scholars have initiated AI-related research since the end of the 20th century,

This paper presents a mathematical allocation model for the opening and assignment of distribution centers, considering constraints such as capacity. The model includes objective functions to minimize total distance, minimize return costs, maximize distribution reliability, and maximize the reliability/distance ratio. The results demonstrate a significant improvement in key metrics: the model reduces the total distance traveled by 55% (from 970,760 to 434,918 meters), cuts return costs by 58% (from $9,831,600 to $4,151,500), and enhances delivery reliability by 10% (from 73.5% to 83.3%). Implementing the model improves logistics strategy by increasing customer delivery reliability and reducing product returns due to delays or unmet specifications.

En este estudio, se ha investigado y evaluado el uso de un modelo de red neuronal recurrente (RNN) del tipo Long Short-Term Memory (LSTM) para la gestión energética en el hogar (HEMS). Utilizando datos de consumo de energía de diferentes dispositivos en intervalos de 15 minutos, se implementó un modelo LSTM para predecir la conexión de dispositivos y estimar su consumo de energía. Los resultados obtenidos mostraron que el modelo LSTM logró una precisión satisfactoria en la clasificación de la conexión de dispositivos y una estimación precisa del consumo de energía. Estos hallazgos destacan el potencial del enfoque propuesto para mejorar la eficiencia energética en los hogares al permitir una gestión más inteligente del consumo de energía de los dispositivos. Además, el modelo proporciona insights valiosos sobre los patrones de consumo de energía, lo que puede ayudar a los usuarios a tomar decisiones informadas para reducir su consumo y optimizar el rendimiento de sus dispositivos.

This paper analyzes public Wi-Fi networks distributed throughout Barranquilla. For this purpose, metrics that allow measuring the performance of these networks in terms of service availability, confidentiality, and physical security were identified and evaluated. Additionally, several aspects were considered, from coverage and connection speed to the security of authentication and integrity of the transmitted data. For which free software tools such as x, y, and z were applied to facilitate data retention. The results show the strengths and weaknesses found in each network evaluated regarding vulnerabilities and some deficiencies.This study contributes significantly to the field by providing a framework for evaluating and improving public Wi-Fi networks, thus promoting their safe and efficient adoption in public environments.

This work presents the design of a model based on machine learning to determine the growth of IT capabilities in organizations. The model allows the IT leader to monitor, control, and delineate the technological capabilities of the evaluated organization. The model emphasizes finding the timing and proportion of technology investment within organizations. The structure of the model is based on standards, frameworks, and best practices in information security, risk management, contingency plans, and quality. The results of this research, with a quantitative focus, are validated in transportation sector organizations located in the Colombian Caribbean region. The findings identify the historical data of IT capabilities according to current regulations and the models and standards of each organization as key factors for implementing the model. Also, implementing the model has allowed participating companies to reduce operational costs by 20% by optimizing server capacity and better planning investments in technological infrastructure.