Search

This research develops a model to predict the results of Colombia’s national standardized test for Engineering programs. The research made it possible to forecast each student’s results and thus make decisions on reinforcement strategies to improve student performance. Therefore, a Learning Analytics approach based on three stages was developed: first, analysis and debugging of the database; second, multivariate analysis; and third, machine learning techniques. The results show an association between the performance levels in the Highschool test and the university test results. In addition, the machine learning algorithm that adequately fits the research problem is the Generalized Linear Network Model. For the training stage, the results of the model in Accuracy, AUC, Sensitivity, and Specificity were 0.810, 0.820, 0.813, and 0.827, respectively; in the evaluation stage, the results of the model in Accuracy, AUC, Sensitivity, and Specificity were 0.820, 0.820, 0.827 and 0.813 respectively.

This research aims to design a quality management tool in education. The methodology comprises two stages: first, the construction of the decision tree model, and second, the efficiency evaluation. For the validation and development of this research, the data modelled corresponds to the standardized exams for higher education in Colombia of ninety industrial engineering degrees. Among the results, the citizenship skills (CC_PRO) generate the most significant contribution to the model. On the other hand, the written communication competence (CE_PRO) generates a minor contribution to the model. In addition, the most relevant result of the research is the design and validation of a tool to estimate educational efficiency using the efficiency analysis tree (EAT) and data enveloping analysis (DEA) models. The proposed tool allows the generation of specific targets to increase the level of efficiency of universities through the nodes of the decision tree, which contributes to the spectrum of knowledge on models for educational management. In conclusion, this research presents a tool for the management of educational processes through the analysis of efficiency using EAT, estimating the efficiency of universities and setting the foundations for forecasting future efficiency scenarios. © 2023 Institute of Advanced Engineering and Science. All rights reserved.

Las instituciones universitarias se enfrentan a retos de competitividad que se enmarcan en las exigencias de la sociedad, los cambios de preferencia de los estudiantes y la evolución tecnológica; esto plantea la necesidad de generar herramientas objetivas para la toma de decisiones que permitan identificar las instituciones que están aportando efectivamente a la formación de profesionales de alto nivel. En el presente trabajo se desarrolla una metodología bajo el enfoque Seis Sigma para medir el desempeño del sistema educativo. Ahora, para evaluar la metodología se utilizan datos del examen colombiano para la evaluación de la calidad en la educación superior para el área de Ingeniería. Los resultados muestran que el nivel sigma del sistema educativo colombiano se encuentra en Z= 2,17 y un rendimiento Y= 75%, lo cual dentro de los criterios establecidos en la investigación es considerado un nivel Aceptable. Como aporte significativo del estudio se presenta una estructura para la evaluación, monitoreo y mejoramiento de la de educación universitaria utilizando el enfoque Seis Sigma. La estructura propuesta es replicable y reproducible a cualquier entorno educativo donde se implementen exámenes estandarizados.

The disappearance of people is a multidimensional phenomenon, in which several aspects must be considered. It affects people’s security perception and consumes police resources in its treatment. Therefore, does exists an emotional circumstance for the relatives of the missing person. At the same, the police departments must develop a search task, in most cases with much uncertainty. In this research, a predictive model to predict missing people’s status is presented. The information used to create the model come from the Colombian legal Medicine Institute, in a public dataset composed of 6202 cases and 11 variables. The output variable was the final disappearance status, with the categories Appears Dead, Appears Alive, and Still Disappeared. Three supervised machine-learning algorithms were trained and tested for the model creation, K-Nearest Neighbours, Decision Trees, and Random Forest. The study was divided into three phases, first considering all the output categories. In the second phase, generating a binary classification for the Appeared and Not appeared instance. Thirdly, models were built to predict the status of appeared persons, Appears Alive or Appears Dead. The K-NN algorithm outperforms the other models with an Area under the curve value of 94.8%. © 2021 Romanian Society of Legal Medicine.

This project formulates a methodology that combines Principal Component Analysis (PCA) and Data Envelopment Analysis (DEA) to perform the analysis of the financial efficiency of the tour operators of the San Andres archipelago in Colombia of the year 2018. Through this methodology it is intended to generate an exhaustive analysis of efficiency that has as inputs in the model the number of workers, operational costs, current assets, operating revenues; and as outputs there is the operating income and the net income. In addition, a Benchmarking analysis will be designed which generates a ranking of the most efficient to the least efficient companies, and finally, a report of the objectives that a unit of study must meet in order to be efficient, the latter to identify the strengths and weaknesses of efficient and non-efficient units. This research concludes that 8 (36%) of the companies in the study are efficient while (14) 64% are inefficient, it was also possible to identify that improvement opportunities can be focused on reducing their operational costs., En este proyecto se formula una metodología que combina el Análisis de Componentes Principales (PCA, por sus siglas en ingles) y Análisis Envolvente de Datos (DEA, por sus siglas en ingles) para realizar el análisis de la eficiencia financiera de los operadores turísticos del archipiélago de San Andrés en Colombia del año 2018. A través de esta metodología se pretende generar un análisis exhaustivo de eficiencia que tenga como entradas en el modelo el número de trabajadores, costos operacionales, activos corrientes, ingresos operacionales; y como salidas se tiene la utilidad operacional y la utilidad neta. Además, se va a diseñar un análisis de Benchmarking el cual genera un ranking de las empresas más eficientes a las menos eficientes, y finalmente, se entrega un reporte de los objetivos que debe cumplir una unidad de estudio para que pueda ser eficiente, esto último para identificar las fortalezas y debilidades de las unidades eficientes y no eficientes. Esta investigación se concluye que el 8 (36%) de las empresas del estudio son eficientes mientras que el (14) 64% son ineficientes, también se logró identificar que las oportunidades de mejora se pueden enfocar en la reducción de sus costos operacionales de las empresas.

The present research is about the analysis financial efficiency of Colombian companies based on the city of Cali certified by the BASC label, for this purpose we used the linear programming technique called Data Envelopment Analysis (DEA), applying the CCR-O routine aimed to outputs. As input variables, it was worked with: Subtotal of inventory Total Current Assets, Plant and Equipment property and Suppliers, and as output variable, Operating Income. The quality of this work is based on the use of primary information collected by the Superintendence of Corporations in 2014. As results we find that average efficiency of 42 companies under study was 33.95%, besides only five companies reached highest efficiency levels. JEL CLASSIFICATION L69, RESUMO Esta pesquisa desenvolve uma análise de eficiência de empresas colombianas localizadas na cidade de Cali, certificadas no selo BASC. Para tanto, utilizou-se a técnica de programação linear denominada Data Envelopment Analysis (DEA), aplicando a metodologia CCR-O focada nas saídas. Como variáveis de entrada, trabalhamos com: Sub-estoque total, Ativo total atual, propriedades da planta, equipamentos e fornecedores. Como variável de saída, foi utilizado o lucro operacional. A qualidade deste trabalho é dada pelo uso de informações primárias coletadas pela Superintendência de Empresas em 2014. Como resultado, verificou-se que a eficiência média das 42 empresas pesquisadas foi de 33,95%, e apenas cinco empresas atingiram altos níveis de eficiência. CLASSIFICAÇÕES JEL L69, RESUMEN La presente investigación desarrolla un análisis de eficiencia de empresas colombianas localizadas en la ciudad de Cali, certificadas en el sello BASC. Para este propósito se utilizó la técnica de programación lineal llamada Análisis Envolvente de Datos (DEA), aplicando la metodología CCR-O enfocada a las salidas. Como variables de entradas, se trabajó con: Sub-inventario total, Activos totales actuales, plantas propiedades, equipos y proveedores. Como variable de salida se utilizó el ingreso operativo. La calidad de este trabajo está dada por el uso de información primaria recolectada por la Superintendencia de Sociedades en 2014. Como resultados se encontró que la eficiencia promedio de las 42 empresas en estudio fue del 33.95%, además solo cinco empresas alcanzaron altos niveles de eficiencia. CÓDIGOS JEL L69