Maestría en Matemática Aplicada
Permanent URI for this collectionhttp://repositorio.uta.edu.ec/handle/123456789/32203
Browse
Item Aplicación de algoritmos de Machine Learning para predecir la deserción estudiantil en alumnos de primer y segundo semestre en universidades públicas del Ecuador.(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2023) Rodríguez Vásconez, Cristóbal Alejandro; Benalcázar Palacios, Marco EnriqueSe estima que en Ecuador la tasa de deserción en los dos primeros semestres de universidad es del 20%. Existen factores socioeconómicos que influyen en el abandono académico de un estudiante. La carencia de programas que atiendan la insatisfacción estudiantil provoca que no se detecten problemas a tiempo y no se puedan aplicar acciones correctivas oportunamente. En este proyecto se aplican técnicas de Machine Learning para predecir la deserción estudiantil a partir de factores seleccionados: socioeconómicos, psicológicos, demográficos y académicos. Partimos de la recolección y tratamiento de datos y se usaron Redes Neuronales Artificiales para crear un modelo que clasifica a un estudiante entre desertor o a salvo de deserción. Se evalúan las métricas Acurracy, sensibilidad y especificidad para determinar qué tan eficiente es el modelo. El modelo final es capaz de clasificar estudiantes a salvo de deserción de forma correcta el 87% de las veces y logra clasificar a desertores de forma correcta el 60% de las veces.Item Aplicación de una modelo estocástico para el análisis RAM de máquinas rotatorias en la Industria 4.0(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría Aplicada, 2023) Zambrano Valverde, Tatiana Paola; Ponsot Balaguer, Ernesto AntonioLa aplicación de conceptos de la Industria 4.0 promovidas a través del mantenimiento predictivo de un activo industrial, marca la forma de la gestión operativa de una fábrica a largo plazo. El an´alisis de la data histórica de los activos, brinda la oportunidad de aplicar técnicas como el modelamiento de datos, que definen el comportamiento de las máquinas a través del tiempo. Este paper presenta un análisis de confiabilidad, disponibilidad y mantenibilidad (RAM) de un grupo de ventiladores industriales que forman parte de un proceso de fabricación de clinker, desde una perspectiva que relaciona datos hist´oricos de vibración, con los estados que toman las máquinas clasificados seg´un el estándar ISO 14694. Para ello, se caracterizan series temporales por cada ventilador, y se obtienen m´etricas descriptivas, que habilitan la aplicación de un tipo de modelo autorregresivo integrado de media móvil, para predecir las condiciones que tomarían los equipos en los siguientes doce meses asociándolas al cálculo de los indicadores RAM, que definen la toma de decisiones en el marco operativo de la planta.Item Diseño de un modelo matemático para estimar la deserción estudiantil mediante técnicas de análisis multivariado en una institución de educación superior tecnológica(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Vinueza López, Cristina Nataly; Loza Aguirre, Edison FernandoEXECUTIVE SUMMARY In this research, a logistic regression model was used to estimate student dropout from the IST Luis A. Martínez Agronómico. The data of 849 students registered in the institute between 2018 and 2020 was used to build the model. The independent variables considered for the model were: gender, marital status, age, career, repetition, occupation and economic status. We used the KDD methodology to estimate the mathematical model, which allows generating information from a database with the records to be studied. In the evaluated period, 82.45 percent of the students did not dropout but 17.55 percent did it. In the study, four logistic regression models were established, the first one includes all the independent variables but only the ‘career’ variable was significant. The ‘age’ and ‘gender’ variables were eliminated (higher p-value) for generating a second logistic regression model, where the ‘repetition’ and ‘career’ variables were considered significant. Subsequently, the highest p-value variables, ‘marital status’ and ‘economic status’ were eliminated for obtaining a third logistic regression model wherein the ‘repetition’ and ‘career’ variables were the only significant ones. Finally, it was chosen the logistic regression model 4, which only includes the career and repetition variables as the only significant ones. The null hypothesis was rejected because the coefficients Beta 1 and Beta 2 of the variables ‘career’ and ‘repetition’ aren´t zero. The logistic regression model 4 correctly classified 83 percent of the training data and 79 percent of the test data. Additionally, we build a prediction model based on decision trees, which established ‘career’ as a unique explanatory variable. The F1_Score value of the logistic regression model 4 was higher than the F1_Score value of the decision tree model.Item Diseño de un modelo matemático para optimizar las rutas de recorrido del proceso de recolección de desechos sólidos para el Cantón Valencia(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Escudero Andino, Flavio Fernando; Ponsot Balaguer, Ernesto AntonioExecutive Summary The optimal urban waste collection route for the canton of Valencia in the province of Los Ríos, Ecuador is presented. A version of the integer linear programming model is selected, known as the “traveling salesman problem”, in which the start and return points do not necessarily coincide. A prior quantitative and graphic analysis of the available data is carried out using the program for geographic information systems QGis and the Pandas library for data management in Python. A strategy is discussed in which the sector under study is divided into five collection groups, made up according to the algorithm known as “k-Means ”. The proposed solution shows to be adjusted to reality and, although it does not imply a substantive decrease in intra-urban routes, since the scale of the problem is very small, it does produce a decrease in operating costs of the order of 18.15 percent, compared with the current situation.Item Diseño de un modelo para el control del consumo de filtros y lubricantes del equipo caminero y maquinaria pesada del GAD del Cantón La Maná mediante algoritmos de inteligencia artificial(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2023) Ortiz Reyes, Juan Carlos; Loza Aguirre, Edison FernandoEl objetivo principal de este estudio fue dise˜nar un modelo matemático utilizando algoritmos de inteligencia artificial para predecir el consumo de lubricantes en la maquinaria pesada del GAD del Cantón La Maná. Para ello se recopilo información de diferentes tipos de algoritmos de inteligencia artificial que podrían ser útiles para la predicción de consumo, y se eligió la red neuronal artificial no lineal autorregresiva como la mejor opción. La información utilizada en este estudio fue obtenida de los registros diarios de la Unidad de Transporte y Maquinaria del GAD Municipal La Maná en los años 2018 y 2019, donde se registraron los kilometrajes y horómetros diarios al inicio y al final de la jornada laboral. La precisión del modelo se evaluó mediante el cálculo del error medio cuadr´atico (MSE), que mide la diferencia cuadrática promedio entre los valores predichos y los valores reales. Los resultados mostraron que la red neuronal que utiliza el algoritmo Retropropagación Levenberg-Marquardt con la arquitectura 128, 64 en las capas ocultas de la red neuronal fue el mejor modelo, con un MSE de 0.000301. En resumen, se concluye que el modelo de red neuronal no lineal autorregresiva puede ser una herramienta ´util para predecir el kilometraje y hor´ometro de las maquinaria pesada, y por ende estimar el consumo de lubricantes, lo que podría permitir una mejor planificación y optimizaci´on del uso de lubricantes.Item Elaboración de un modelo matemático de identificación de seguridad alimentaria en el contexto de la pandemia por covid-19 en la provincia de Tungurahua.(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2023) López Silva, Victoria Monserrath; Gavidia García, José LuisLa seguridad alimentaria a nivel mundial es un tema preocupante debido a que afecta el acceso a los alimentos por parte de los hogares. La pandemia originada por el virus SARS-CoV-2 a finales del año 2019 requirió el distanciamiento y aislamiento social, esto, representó para los países conflictos ambientales, sociales y económicos. En concreto, surgieron problemas laborales que generaron el aumento del desempleo y la pobreza. Por tal motivo, en esta investigación se elaboró un modelo matemático usando regresión logística para identificar las variables influyentes de la seguridad alimentaria en el contexto de la pandemia por Covid-19 en la Provincia de Tungurahua y se analizó la diversidad alimentaria antes y después del evento. Se encontró diferencias significativas en el grupo de alimentos constituido por raíces y tubérculos, frutas, carnes, huevos, pescados y mariscos, leches y productos lácteos, leguminosas y semillas, aceites y grasas, dulces y especies, condimentos y bebidas. No se observaron cambios en el consumo de cereales, verduras y huevos. El modelo arrojó como variables influyentes área urbana, género masculino, nivel de escolaridad primaria, secundaria y universidad, nivel de ingresos mayor a $1200, gastos en alimentación menor a 200 y número de personas que viven en el hogar (PVH). En la provincia de Tungurahua para los hogares de la muestra bajo estudio se observó que durante la pandemia por COVID-19 predominó la inseguridad alimentaria leve.Item Estudio de los modelos de regresión paramétricos polinomiales y modelos de regresión no paramétricos B-Splines. Aplicaciones en ingeniería(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Toalombo Rojas, Byron Miguel; Meneses Freire, Manuel AntonioThe study of polynomial parametric regression models and non-parametric B-Splines regression models is carried out based on particular applications in engineering. The cases considered are a car against the bus bodywork crash simulation as part of the structural design and the relationship of the climatological variables in the weather station of San Antonio de Pichincha. A non-experimental cross-sectional methodological design is made, being a correlational type of research. The appropriate regression models for each relationship are established with the use of R software and taking into account the criteria: rejection of the nullity of the coefficients of the models by Student's t-hypothesis test, the validity of the models by Snedecor's F test of the ANOVA table, the goodness of fit, 95% confidence intervals, and compliance with the assumptions of normal distribution, no autocorrelation, and homoscedasticity of the residuals for the polynomial regression (Shapiro-Wilk, Kolmogorov-Smirnov corrected by Lilliefors, Durbin-Watson and Breusch-Pagan tests, respectively). The Wilcoxon nonparametric test was used to select the most suitable regression model based on the lengths of the confidence intervals. Based on the results obtained, the polynomial parametric regression models fit well when the curves have a parabolic shape or follow a pattern without abrupt changes in curvature. It means the model can define better to the relationships of the vehicle impact simulation that have the speed of the impacting vehicle as an explanatory variable. In contrast, the nonparametric Bsplines regression models provide a better fit when the curves are bell-shaped with more abrupt curvature changes.This model can adapt better to the conditions of the climatological variables as a function of the time of dayItem Implementación de un sistema inteligente para la identificación vehicular(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Cáceres Mayorga, Paúl Alejandro; Reinoso Astudillo, Cristina IsabelThe main objective of this research work was to implement an intelligent system capable of classifying vehicle and automotive license plates, as well as self-correcting recognition errors, for which it was based on the design of an algorithm capable of detecting and classifying vehicles, implementing artificial intelligence. Once the process to be followed was identified, a source code was implemented for the detection of the types of license plates using the convolutional neural network WPOD in which the data of the edge, width and height of the plate were specified so that it only provides the photo of license plate. For the binarization process used in the research, the Otsu algorithm was used, which converts the images into the gray, blur, binary and dilation scales, applying filters that can obtain the location segments of the letters and numbers. Finally, an effective system was obtained, with acceptable detection capacity, since design parameters of the architecture of each type of red were established, which achieved a satisfactory solution to the problem of identification, classification and validation of the characters of Ecuadorian license plates.Item Implementación de un sistema predictivo con redes neuronales para el control del comportamiento de la planta Festo MPS-PA(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Soria Mejía, Daysi Maribel; Benalcázar Palacios, Freddy GeovannyIn the present research work, the implementation of a Dynamic Matrix Predictive Controller (DMC) with Neural Networks was carried out for the level control process (liquid) of a Festo Compact Workstation plant of the hydraulics and pneumatics laboratory of the Technical University of Ambato. The dynamics of the plant was found through the training of a feedforward neural network, the training and testing data used were obtained by conducting an experiment that consists of applying different step inputs to the plant and the response of the system to said input. The algorithm implemented was that of a dynamic matrix predictive controller, for which it is necessary to know the mathematical model of the level process represented as a transfer function, said mathematical model was built using the exponential regression method by least squares.Item Modelación matemática para un control robusto de la planta Festo MPS-PA Compact Workstation mediante la normativa IEC-61499(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Bustos Pulluquitin, Sergio Patricio; García Sánchez, Marcelo VladimirIn the present research work, the implementation of a GPC control with restrictions was carried out in a FESTO MPS-PA CompactWorkstation level plant through the modeling of function blocks, thus obtaining an industrial control application with decentralized logic which is one of the objectives of the IEC 61499 standard. The IDE used compatible with IEC 61499 was the 4DIAC IDE and the runtime environment (runtime) 4DIAC FORTE, in addition the BBB card was used for reading and writing data from sensors and actuators respectively. The mathematical model of the level plant represented as a transfer function was found by two methods: the first using linear differential equations and the second using experimental data of the response to the step of a first order system. Three types of restrictions were implemented: in the control increment ( Delta u(t)) to avoid unnecessary efforts in the actuator and the output restrictions (y(t)) and control (u(t)) that will prevent damage to the plant, the GQPA algorithm was used to optimize the objective function.Item Modelado matemático predictivo para la sostenibilidad de los emprendimientos productivos(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Luzuriaga Jaramillo, Héctor Alberto; Ponsot Balaguer, Ernesto AntonioExecutive Summary Models that explain and predict failure and success are very valuable to entrepreneurs in the production sector. The implementation of an explanatory and predictive mathematical - statistical model is presented for the study of the sustainability of productive enterprises in the Cevallos, Quero, Mocha, Patate and Tisaleo cantons of the province of Tungurahua, Ecuador. A logistic regression model is proposed and adjusted on a sample of 1546 entrepreneurs. The variable selection method is stepwise which, following the parsimonious principle, incorporates only the variables that are significant. The deviance goodness-of-fit test suggests that such a model fits the data appropriately. Several hypotheses are checked and discussed, the most important result being the demonstration that there are factors that can be used to explain the success of productive entrepreneurs. These are: Instruction, time of the undertaking, staff, canton and if it has been suspended.Item Modelo de Predicción de Deserción Escolar en los Estudiantes la Unidad Educativa “Los Andes”(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2023) Vásconez Altamirano, Gladys Edilma; Meneses Freire, Manuel AntonioLa deserción escolar ha sido un problema que ha existido a lo largo de la historia tanto a nivel mundial como en el Ecuador, sin embargo, desde el inicio de la pandemia por COVID-19 ha tenido un incremento inédito, el mismo que se ha dado en todas las instituciones educativas del país, esta pandemia ha tenido un impacto severo sobre las distintas variables socioeconómicas en los estudiantes y todo su entorno, las instituciones educativas tuvieron que cerrar sus puertas a la educación presencial generando una nueva modalidad de aprendizaje que es la enseñanza virtual, haciendo uso de distintas plataformas educativas y diversos recursos tecnológicos lo cual tuvo un efecto negativo provocando un alto índice de deserción estudiantil. Según El Ministerio de Educación del Ecuador hasta noviembre de 2021 unos 150.000 niños y adolescentes han dejado de estudiar producto de la crisis ocasionada por la pandemia del COVID-19. Este trabajo de investigación presenta varios aportes que ayudan a dar respuesta a varios de los problemas que surgen en la educación producto de la pandemia por COVID -19, la propuesta base estará en diseñar un modelo predictivo que ayude a estimar los índices de deserción de estudiantes en la unidad educativa Los Andes. Al estudiar los efectos que trajo consigo la pandemia se tiene que ha producido alteraciones de orden psicológico en los estudiantes, la falta de recursos económicos para acceder a las clases y la falta de tecnología ha hecho que varios jóvenes sientan un desgaste emocional, falta de concentración y desmotivación, haciendo que contemplen la idea de dejar sus estudios. A continuación, se realiza una descripción del proyecto de investigación, refiriéndonos primero al modelo matemático que se propone efectuar y la documentación que lo sustenta, a las pruebas empíricas que se propone realizar y a la metodología a aplicar. Se específica, el tipo de análisis a efectuar, así como los datos sobre los cuales se aplica un modelo mediante la técnica de mínimos cuadrados, determinando resultados obtenidos y la discusión de estos. Finalmente se presenta la propuesta de solución.Item Modelo de predicción de riesgos psicosociales en el transporte urbano de pasajeros usando técnicas de Inteligencia Artificial(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Lara Satán, Amado Antonio; Loza Aguirre, Edison FernandoEXECUTIVE SUMMARY The city bus driver job ranks among the highest risk and most stressful modern occupations. Modern technologies provides greater autonomy and work flexibility, however they also expose drivers to psychosocial risks, which leads to work stress. Consequently, the early prediction of stress and their associated risk, would contribute to make preventive decisions. The objective of this study is to develop a model that allows predicting psychosocial risks in urban passenger transport in the city of Ambato, applying supervised machine learning techniques. For this purpose, we used data set of occupational psychosocial risk of urban bus drivers obtained with the Fpsico 4.0 questionnaire. The study applies the methodology for the identification, analysis, and evaluation of psychosocial risks of the INSHT of Spain and the Cross Industry Standard Process for Data Mining (CRISP-DM) framework. The classification is performed with the three non-parametric supervised algorithms: k-nearest neighbors, decision trees and support vector machine. The evaluation metrics of the algorithms used are the Jaccard index and F1-score. The experimental results show that the support vector machine model shows better performance with an F1 score of 93 percent and the Jaccard score of 87 percent.Item Modelo de predicción del resultado en exámenes de acceso a la educación superior para estudiantes que se preparan en centros de capacitación preuniversitaria usando algoritmos de Machine Learning.(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2023) Coba Gavilánez, Christian Danilo; Benalcázar Palacios, Marco EnriqueLa nota de admisión para el ingreso a la educación superior define si un estudiante ingresa o no la carrera de su interés. En Ecuador se ofertan cupos para el 56% de los postulantes a tercer nivel [1]. Esto hace que los estudiantes que optan por un cupo se preparen adicionalmente en un programa preuniversitario. Los cursos de preparación preuniversitaria tienen la misión de hacer que un estudiante obtenga una buena nota y pueda postular para tener una mayor probabilidad de ingreso a la universidad. Usualmente un programa preuniversitario consta de varios procesos académicos y evaluaciones continuas. En este trabajo se propone tener una predicción de la nota que sacará un estudiante en su examen de ingreso a la universidad antes de completar el programa preuniversitario. Adicionalmente se desea conocer cuáles son los factores de mayor relevancia que hacen que esta nota varíe. En los resultados se puede ver que la filial Ambato, un curso de 10 meses y los simulacros de exámenes son factores que tienen un impacto directo en la nota final de admisión. Los modelos de predicción implementados en este trabajo se basan en el uso de regresión lineal y redes neuronales artificiales (RNA). Los resultados de predicción de ambos modelos son similares, pero la ventaja del modelo de regresión lineal es que se puede interpretar cada una de las variables predictoras. Los datos y las variables de interés se obtuvieron del centro de estudios Quality Up, con información de procesos de admisión de 300 estudiantes pertenecientes al ciclo sierra 2022.Item Modelo estadístico para la planificación de la producción en una ensambladora de vehículos, mediante la metodología Lean Manufacturing(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Gavidia García, José Luis; Meneses Freire, Manuel AntonioEXECUTIVE SUMMARY The main objective of this study was to design a statistical model for production planning in a vehicle assembly plant, using the Lean Manufacturing methodology, which covers topics related to the calculation of Takt-Time, focused on the analysis of times and movements, with in order to eliminate downtime, thereby generating an adequate balancing of production lines that are coupled with production planning in the assembly of the HAVAL M4 AC 1.5 5P 4X2 Model vehicle produced by the company Ciauto Cía. Ltda. Initially, the identification of the different assembly processes that are executed in the ten workstations was carried out, in order to carry out the taking of times through the use of a stopwatch, managing to determine the bottleneck within the assembly process of the Model M4, condescending of this condition to manage alternative solutions for a production of eleven units per day. In the analysis of the Installed Capacity (CI) a value of 12 vehicles / day was obtained, this means that the assembler based on its resources, equipment and infrastructure is capable of assembling 6 M4 Models daily, this value confirms that in addition to the Compliance with the production plan set by the assembly coordinator, each batch of M4 corresponding to 30 cars can be assembled in approximately one week and one day, systematically satisfying the demand of the national market. The data from the investigative work highlight the need to provide a statistical tool that guides efficiently and sustainably in the short, medium and long term the planning of production in the assembly area of the company Ciato Cía. Ltda., In order to generate efficient management of resources in a vehicle assembly company.Item Modelo estadístico para valorar las líneas de transporte público de pasajeros en la ciudad de Riobamba(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Londo Yachambáy, Fabián Patricio; Meneses Freire, Manuel AntonioEXECUTIVE ABSTRACT Public passenger transport is considered an inefficient service in Ecuador and therefore in the city of Riobamba, although there have been attempts on several occasions to achieve high levels of safety, comfort and quality, there is almost nothing that could be improved. Faced with this problem, it is about applying a statistical model that allows me to assess public passenger transport, for which an analysis was carried out in three axes that involve this means of transport, such as Infrastructure, ascent and descent of passengers and quality of service applying the survey technique and use of a questionnaire to obtain information, with which variables were obtained that helped to build my Multinomial Logit model.Item Modelo matemático basado en las técnicas de detección de bordes y propagación de texturas para restaurar imágenes fotográficas documentales RGB(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Jordán Bolaños, Carlos Rodrigo; Gordón Gallegos, Carlos DiegoEXECUTIVE SUMMARY The restoration of RGB documentary photographic images, using a mathematical model based on two fundamental techniques texture propagation and edge detection seek to give balanced priority to the two techniques, that is, neither should prevail over the other, using the notation for image restoration: I = Image, ɸ = source region. It is the known region of the image, Ω = target region. It is the region to restore and 𝛿Ω = border. RGB (color) documentary photographic images focus on people and social groups, to show aspects of their daily lives and are born with the intention of capturing reality in a still image. Edge detection is a technique that facilitates the identification of image objects and therefore their recognition. If the purpose of edge detection algorithms is to obtain as a result an image where the pixels of those points of the original image where sudden changes in intensity are presented are highlighted. In this respect it is worth highlighting, among others, the techniques related to zero crossing (“Zero crossing”) and level set (“level set”). On the other hand, texture propagation is a technique that consists of building from a small sample image, a large image that preserves the structure of the sample. An important part of the model that is proposed in this work, consists in filling the image by pixels, first of all, you must count on the area to be restored is well defined so that the neighboring textures maintain balance and harmony in the final result. The proposed mathematical model is designed to obtain from the original image, first the gradient of the image followed by the extraction map which is nothing more than the silhouette of the object to be extracted; then the energy map of the image is created to proceed with the removal of seams. Seam after seam will be removed until you create a resulting image similar in size and characteristics to the original. Finally the restoration will be done considering the edges and textures of the original image, with this it will be evaluated if the restored documentary photographic image is visually satisfactory. The similarity criterion to make the comparison between the original and resulting image has to be of mathematical type, that is, an algorithm will compare pixel by pixel the two images showing on screen how efficient the algorithm is. Under the above context it is said that the restoration of images through mathematical modelling seeks to refine the amendment in an intelligent way, that is to say, that at the end of the restore process, the image is visually pleasing to the observer, where the subtraction of elements or the size modification does not harm the harmony of colors, shapes and textures.Item Modelo matemático de inventarios en la cadena de suministros y su impacto en la comercialización de productos de fibra acrílica, método Multicriterio ABC(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Toscano Guerrero, Francisco Eduardo; Jurado Lozada, Marco AntonioABSTRACT Companies worldwide today compete within the framework of technology and the resources they have, so that the market is always satisfied according to their needs and preferences, it is for this reason that at the national level and punctually at the level Local strategies are established which allow companies to develop in a better way, incorporating in their policies new elements that lead to a better development in the field of product marketing. The problems that affect companies daily make them look for alternative solutions, specifically speaking of inventories as an important capital investment found in assets which entail expenses of materials, administrative personnel, and storage warehouse, which consequently need of practical solutions that allow a balance with the customer and product rotation. The main objective of this study leads to the solution of this problem by incorporating a mathematical model incorporated with a computational application that allows finding solutions to the problem of ordering the products that are in greatest demand in the local and national market, optimizing resources and obtaining a balanced inventory with respect to the rotation of products and managing to raise its level of marketing based on sales. The computational application is a very versatile and practical tool, managing to analyze many elements of various criteria, accelerating the process of selecting the best products based on each of the characteristics that the market prefers, thus maximizing the impact of the marketing of acrylic fiber products from the company "Lanas Elsy". Vilfredo Pareto's ABC classification technique is used in many analyzes at all levels, becoming one of the best alternatives for developing inventories worldwide, combined with the multi-criteria mathematical model and the neglected computational application in C # and Matlab, are part of a very important input for the development of the company.Item Modelo matemático de la producción de la empresa Salinerito en la Provincia Bolívar(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Sánchez Verdezoto, Carlos Alfredo; Benalcázar Palacios, Freddy GeovannyEXECUTIVE SUMMARY The present research work is descriptive, predictive and longitudinal with a trend. The objective was to develop a mathematical model for the production of the company Confites El Salinerito in the Bolívar province, from the production values in the period from January 2017 to July 2020, the data were obtained directly from the company. To achieve the research objectives, the methodology proposed by Box-Jenkins was applied, which describes the characteristics of the time series in terms of trend, seasonality and stationarity; The free software RStudio version 4.0.1 was applied for the estimation of the parameters, the processing and the analysis of the data. It was concluded that the mathematical model that most accurately adjusts to the production values of the company was SARIMA (1, 1, 1) (1, 1, 1) 12, the same one that allowed making the monthly production forecasts for the period from August 2020 to January 2021. The results obtained by the developed model and the methodology used by the company to establish its monthly production were also compared.Item Modelo matemático de optimización para planeación de redes FTTH a través de programación estocástica multietapa(Universidad Técnica de Ambato. Facultad de Ingeniería en Sistemas, Electrónica e Industrial. Maestría en Matemática Aplicada, 2021) Cunalata Landa, Myriam Paola; Salazar Escobar, Fabián RodrigoEXECUTIVE SUMMARY In this research project a mathematical optimization model is used for planning FTTH networks through multistage stochastic programming, the method of reduction and construction of scene trees is studied in order to choose the appropriate modeling approach, in addition to define an optimization model for the projection of FTTH telecommunications networks, finally the effectiveness of the new data model obtained from the company FASTER ISP is evaluated. With the mixed methodology used in this project, quantitative and qualitative data are collected, analyzed and combined that allow to achieve the proposed objectives, to reach these objectives the different concepts of FTTH networks, mathematical models, K-means optimization algorithms are studied, Setcover, Dijkstra, Brownian Movement. The use of these algorithms allows to process the information of the clients, reduce the scenarios to find the optimal one, carry out the stochastic projection of the potential clients and find the input and output variables that are used to define the mathematical model that serves as parameters of the objective function, to optimize various resources such as splitters and the number of fiber optic meters. The algorithms used allow identifying the resources and parameters necessary for optimization, on the other hand, it is evident that the inappropriate location of an equipment splitter in the deployment of FTTH networks leads to oversizing or wasting resources. To define the optimization model, variables and restrictions discovered in the information processing were used, it should be added that the model minimizes the cost of investment in resources. The use of the optimization model reduces the cost of planning in a FTTH network by 39%, on the other hand, future lines of research may be the application of the mathematical model in a WIFI network.