Tesis
Permanent URI for this communityhttps://hdl.handle.net/11285/345119
Colección de Tesis y Trabajos de grado (informe final del proyecto de investigación, tesina, u otro trabajo académico diferente a Tesis, sujeto a la revisión y aceptación de una comisión dictaminadora) presentados por alumnos para obtener un grado académico del Tecnológico de Monterrey.
Para enviar tu trabajo académico al RITEC, puedes consultar este Infográfico con los pasos generales para que tu tesis sea depositada en el RITEC.
Browse
Search Results
- Hyper-heuristic Model Based on Neural Networks for Solving the Metaheuristic Composition Optimisation Problem in Continuous Domains(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-12) Tapia Avitia, José Manuel; Terashima Marín, Hugo; emimmayorquin; Pillay, Nelishia; Ortiz Bayliss, José Carlos; Amaya Contreras, Iván Mauricio; School of Engineering and Sciences; Campus Monterrey; Cruz Duarte, Jorge MarioMetaheuristics (MHs) have been proven to be powerful algorithms for solving highly non-linear and intricate optimisation problems over discrete, continuous, or mixed domains, with applications ranging from basic sciences to applied technologies. Nowadays, the literature is prolific with MHs based on outstanding ideas, but the researchers often recombine elements from other methods. To avoid the frenetic tendency of proposing methods more focused on metaphors than operations, a standard model has been proposed to customise population-based MHs, which uses simple heuristics or search operators extracted from well-known metaheuristics. The framework corresponding to this model can be found as Customising Optimising Metaheuristic via Hyper-heuristic Search (CUSTOMHyS), which facilitates implementing models that explore a heuristic space. Still, they are limited by the nature of the metaheuristics used in such models, as such algorithms does not consider the information gained from previous explorations to enhance the tailoring process. A field of action and improvement that has not been explored is the model implementation to take advantage of previous results and learns from them to boost the performance of the tailoring process. For that reason, we propose a hyper-heuristic model based on neural networks, which is trained with processed sequences of heuristics to identify patterns that one can use for generating modified metaheuristics. Being more specific, the task assigned to the neural network is to predict the simple heuristic from the collection or heuristic space to apply next, considering a sequence of heuristics already applied to the low-level problem. Using the neural networks, the challenge is to define how to generate metaheuristics with a high performance for tackling a family of optimisation problems. This research work propose a novel methodology that decomposes the metaheuristics into several subsequences of heuristics to train the neural network models. To prove the feasibility of the proposed model and training methodology, it is compared against generic well-known basic metaheuristics and other heuristic-based approaches, such as the unfolded MHs. The results evidence that the proposed model outperform an average of 86% of all scenarios; in particular, 91% of basic and 81% of unfolded approaches. Plus, it is worth to highlight the configurable capability of the proposed model: several experiments are carried out to explore a few control variables and show their effects in the model. It proves to be exceptionally versatile regarding the computational budget. After exploring and finding a suitable configuration, we perform an extended analysis of the training computational cost, and a study of the metaheuristics generated by the model. Moreover, we analyse the usage of previously generated metaheuristics on an unseen problem via a few strategies. The proposed model and its metaheuristics show their adaptation capabilities to unseen problems, proving to be a good alternative for real-world application problems.
- Wave characterization and path loss modelling for 5g wireless communication in indoor light industrial deployments at 2.4ghz and 28ghz frequency bands(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-11-23) López Ramírez, Gustavo Adulfo; Aragón Zavala, Alejandro; puemcuervo, emipsanchez; Vargas Rosales, César; Zaldívar Huerta, Ignacio Enrique; School of Engineering and Science; Campus MonterreyIn a world increasingly focused on the wireless interconnection of devices, it is pertinent to carry out an in-depth investigation of the theory surrounding electromagnetic waves and their propagation mechanisms; only in this way it is possible to carry out an adequate characterization. After the research established as a fundamental basis, two measurement campaigns were carried out in semi-industrial indoor environments, selecting two frequency bands: 2.4GHz and 28GHz, both found within the two frequency sets that comprise the 5G networks to subsequently perform an analysis of the data collected and adjust them to the single-frequency floating-intercept path loss model. The application of the empirical model fits the collected data and shows reliability to predict the path loss behavior for both frequency bands; however, specific studies must be carried out to analyze in detail the specific contributions of the propagation mechanisms. The model proves not to be deterministic, but instead presents random variables and must be approached stochastically, in addition to being highly dependent on the analyzed site.
- Rediseño de conector automotriz sellado para erradicar problemas de configuración por ponchado(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2023-03-24) Villarreal Gómez, Sergio Andrés; Tamayo Enriquez, Francisco; emimmayorquin; Hernández Luna, Alberto; Escuela de Ingeniería y Ciencias; Campus Monterrey; Yabur Pacheco, PedroEste documento muestra el proceso de desarrollo de un nuevo diseño para conectores eléctricos de bajo voltaje que utilicen un sello multicavidad y requieran una configuración de cavidades bloqueadas. El problema resuelto son dos defectos que aparecen en los retenedores ponchables y le cuestan decenas de miles de dólares a Aptiv en costos de pobre calidad: Vestigios de ponchado, interferencias plásticas que impiden la inserción de las terminales, y Daños a los postes de sellado, por la fragilidad de los postes. El nuevo diseño plantea el uso de postes de sellado que sean insertados mediante un ajuste de interferencia en el conector, de tal forma que se bloqueen ciertas cavidades para formar la configuración deseada y se eviten los problemas con el diseño que se usa actualmente. Mediante el flujo de desarrollo de productos de Aptiv se plantearon soluciones al diseño con lo que tras el análisis mediante prototipos de impresión Carbon se corroboró el funcionamiento de los postes de sellado como una solución viable para sustituir la tecnología de retenedores ponchables. La aplicación de esta tecnología se puede realizar en conectores que utilicen un sello multicavidad como es el caso del conector 55 vías Ergomate que fue el conector en el que se aplicó este diseño para su evaluación. Este diseño fue catalogado en el sistema de recursos de diseño de Aptiv, se encuentra en proceso de patente y su lanzamiento a producción en masa se mantiene a la espera de fondos para la creación de moldes y herramental.
- Glucose measurement via noninvasive methods(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-12-02) Huerta Ruiz, Samuel Natán; González Hernández, Hugo Gustavo; emipsanchez; Moreno Moreno, Jesús; School of Engineering and Sciences; Campus Ciudad de México; Oliart Ros, AlbertoNoninvasive glucose measurement methods are wide-ranging; they use several different technologies to try and get accurate results. Some try to measure glucose through lacerations on the skin and chemicals, others try to do it analyzing the color of the sclera, and others try to do it analyzing the sweat. For this thesis, a completely noninvasive and chemical free approach is used. Glucose levels are classified into three useful categories (low, medium, and high) trough the use of machine learning and descriptors from chaos theory to obtain a a satisfactory Support Vector Machine (SVM) model. Several classification models are compared by the following metrics: Area Under the Receiver (AUC), accuracy, precision, recall, and their combined information (F1). And lastly, a multipurpose system that uses principles from Internet of Things (IoT) is implemented to integrate a sampling device powered by Arduino with a web app, which in turn uses cloud computing to process data and store it in a remote server to effectively train machine learning models written in Python.
- Implementación de un modelo matemático para el control de espesores de películas de BaTiO3/PVDF por spin- coating(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-11-28) Ochoa Bonilla, Eduardo Ernesto; Sustaita Narváez, Alan Osiris; emipsanchez; Iturbe E K, Jackeline; Escuela de Ingeniería y Ciencias; Campus Monterrey; Ulloa Castillo, Nicolas AntonioLas películas delgadas juegan un papel esencial en el desarrollo e investigación de nanomateriales con propiedades nuevas dando aplicaciones que pueden ser desde un simple recubrimiento para evitar la corrosión, hasta aplicaciones fotovoltaicas y de microelectrónica. Se estudiaron películas sintetizadas a partir de Titanato de Bario (BaTiO3) y Fluoruro de Polivinilideno (PVDF) disuelto en N, N-dimetilformamida (DMF) HCON(CH3)2, por la metodología sol gel para la preparación de las nanopartículas. Las películas fueron depositadas por el método de deposición por centrifugación (spin-coating). Utilizado un modelo matemático propuesto por Emslie et al., se llevó a cabo una modelación para la predicción de grosores y controlar el espesor de las películas a partir de las concentraciones seleccionadas buscando beneficiar la obtención de películas delgadas uniformes y que promuevan la conversión de fase beta cristalina del PVDF. Para la deposición se llevó a cabo un estudio reológico de la solución a envejecer de PVDF / BaTiO3 para conocer algunos parámetros de viscoelasticidad usados en el modelo matemático. Por último, se desarrolló un diseño experimental variando la concentración de nanopartículas añadidas y evaluando los grosores de las películas con las del modelo matemático. Las concentraciones seleccionadas tras las pruebas exploratorias con el PVDF y BaTiO3 variaron de 0.02-0.05 g/10mL y 0.10-0.15 g/100mL respectivamente en relación con el peso del DMF. La velocidad de deposición utilizada fue de 2000 rpm. Las películas se caracterizaron por perfilometría, espectroscopia infrarroja y difracción de rayos X, a fin de conocer la composición y estructura, para el estudio de la conversión de fase β obteniendo hasta en un 80% para las películas con 10% de concentración de BaTiO3 depositadas sobre substratos de silicio y vidrio, evaluando valores similares obtenidos en el modelo matemático para las pruebas usando perfilometría para corroborar el grosor (en µm) de las películas.
- Combination of multiple split evaluation measures in C4.5(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-11) Ramírez Sáyago, Ernesto; Medina Pérez, Miguel Angel; emipsanchez; Sosa Hernández, Victor Adrián; Gutiérrez Rodríguez, Andrés Eduardo; School of Engineering and Sciences; Campus Estado de México; Loyola González, OctavioClassification problems are a key area in Machine Learning. Despite the growing popularity of black-box approaches to solving these problems, there is still a need for solutions that are able to be understood by experts in the field. Among these solutions are Decision Trees, of which C4.5 is a foundational algorithm that continues to serve as the base for subsequent improvements. The literature shows that no single split evaluation measure consistently outperforms every other measure under a wide range of databases. Therefore, one solution to this problem is to combine multiple splitting criteria into a single Decision Tree. Our proposal aims at doing this by using the power of voting methods to aggregate the preferences of multiple splitting criteria to select the best one. We ran a series of experiments to appraise the performance of our proposal. 114 databases were utilized, with different characteristics in terms of the number of attributes, instances, or whether they were binary or multi-class. Two novel splitting criterions found with our proposal were assessed against 11 individual splitting criteria. AUC was chosen as the evaluation criterion for these comparisons, and statistical tests were made to verify statistical significance. Results show that our proposal has a competitive performance against the individual splitting criteria. The Wilcoxon test found that our proposal outperformed the other criteria with statistical significance. On the other hand, The Friedman with Finner post-hoc test ranked our proposal as the best-performing criterion but did not find a statistically significant difference with every individual splitting criterion.
- Coexistence of trust and distrust in technology: a mobile health monitoring systems comparative case study across time(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024-12-01) Cantú Alejandro, Celeste; Güemes Castorena, David; emipsanchez; Elizondo Noriega, Armando; Smith Cornejo, Neale Ricardo; AGuayo Tpellez, Humberto; Escuela de Ingeniería y Ciencias; Campus Monterrey; Cortes Capetillo, Azael JesusEmerging technologies challenge organizations by introducing information systems that transform processes and introduce a series of risks and uncertainties for stakeholders. These conditions make trust and distrust important in the adoption and implementation of technology. Most of the trust in Information Systems literature has focused on building the constructs of trust and distrust between an artifact and the final user, but there is a gap in the analysis of changes in trust and distrust beliefs and their possible coexistence across time. This case study illustrates how trust and distrust manifest across a mobile health monitoring system implementation: the attitudes, objects of trust/distrust, and the changes they exhibit across four stages. The results point at the objects of trust/distrust corresponding to the technology frames that shift during the implementation from an artifact-centric to a more encompassing technology involving other users, programmers, and themselves. The actual technology use leads to changes in perceived risks and expectations, which are fed by the interaction of all users with the system. Behaviors are shaped by coexisting trust and distrust in technology beliefs, which usually compensate for the elements of the technology they distrust. The study concludes that the coexistence of trust and distrust in technology is a constantly shifting phenomenon, leading to a complex yet stabilized use of technology.
- Enhanced mechanical characteristics of Inconel-718 lattice structures produced by laser powder-based fusion with heat treatments(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2022-06-12) Briones Montemayor, María José; Martínez Romero, Oscar; emipsanchez; Elías Zúñiga, Alex; Olvera Trejo, Daniel; Escuela de Ingeniería y Ciencias; Campus Monterrey; Guzmán Nogales, RigobertoThis thesis explores the mechanical properties of four lattice structures—BCC, diamond, IWP, and gyroid—selected for their load-carrying and energy absorption capabilities. Compression tests reveal that the BCC and diamond structures exhibit larger plastic zones, while the IWP demonstrates superior energy absorption per volume, attributed to its smoother surface geometry resembling the BCC structure. The gyroid structure displays the highest yield strength. The selected heat treatments, HT1, HT2, and HT3, yield similar results, though variations in microstructure and grain size significantly affect mechanical properties. HT2, with its δ-phase boundaries, exhibits promising outcomes. The combination of HT2 and gyroid structure enhances yield strength by 94%, making it ideal for high-strength applications. Similarly, the pairing of IWP and HT2 increases energy absorption per volume by 48%, suitable for energy dissipation applications. XRD analysis revealed significant changes in the principal crystalline planes (111, 200, and 220) across different heat treatments. HT2 exhibited the most pronounced phase transformations, with sharp XRD peaks and reduced microstrain, correlating with the largest grain sizes and the highest mechanical performance. HT1 showed initial microstructural adjustments with smaller grain growth and moderate mechanical properties, while HT3 resulted in a balanced microstructure with stable and moderate mechanical performance. Integrating lattice structures with heat treatments enhances material properties, with HT2 emerging as a promising treatment for advanced material applications. Recommendations for future research include exploring microstructural evolution, long-term material performance, fatigue behavior, and cooling rate effects.
- Real-time armed individual detection in video surveillance usingdeep learning and heuristic approaches(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024-12) Amado Garfias, Alonso Javier; Conant Pablos, Santiago Enrique; emipsanchez; Ortiz bayliss, José Carlos; Tarashima Marín Hugo; Gutiérrez Rodríguez, Andrés Eduardo; School of Engineering and Sciences; Campus MonterreyThis researchaimstoenhancetheautomaticidentificationofarmedindividualsinvideo surveillanceinreal-time.Theproposedmethodologyinvolvesthedevelopmentofalgorithms specifically designedforthedetectionofindividualscarryinghandguns,whichincludepistols and revolvers.Toachievethis,theYOLOv4modelhasbeenselectedtodetectindividuals, handguns, andfaces.Subsequently,real-timeinformationisextractedfromtheYOLOmodel, including boundingboxcoordinates,distances,andintersectionareasbetweenhandgunsand individualswithineachvideoframe.Thisinformationfeedsourheuristicsanddifferentma- chine learning(ML)proposed,facilitatingtherecognitionofarmedindividuals.Severalchal- lenges mustbeaddressed,suchasocclusion,concealedguns,andproximityofindividualsto one another.Itencouragesthedevelopmentandcomparisonofdifferenttypesofsolutions. Theyaremadeupofthreeheuristics,seven-armedpeopledetectors(APD),and44APDto use ineachvideoframe(APD4F). The heuristicsaretheDeterministicMethodofCenters(DMC),theDeterministicMethod of Distances(DMD),andtheDeterministicMethodofIntersections(DMI).Furthermore, the APDmodelsareRandomForestClassifier(RFC-APD),MultilayerPerceptron(MLP- APD), k-Nearest-Neighbors(KNN-APD),SupportVectorMachine(SVM-APD),Logistic Regression(LR-APD),NaiveBayes(NB-APD),andGradientBoostingClassifier(GBC- APD). Thereby,IproposetocreateselectorsfordecidingwhichAPDtouseineachvideo frame (APD4F)toimprovethedetectionresults.Besides,weimplementedtwotypesof APD4Fs, onebasedonaRandomForestClassifier(RFC-APD4F)andanotherinaMultilayer Perceptron (MLP-APD4F).Wedeveloped44APD4FscombiningsubsetsofsixAPDs.The most ofAPD4FoutperformedoftheindependentuseofallAPDs.Amultilayerperceptron- based APD4F,whichcombinesanMLP-APD,aNB-APD,andaLR-APD,presentedthebest performance, achievinganaccuracyof95.84%,arecallof99.28%andanF1scoreof96.07%. This researchalsoproposesasolutiontooptimizetheproblemofdetectingarmedpeople when theweaponisnotvisible.Therefore,weapplyrecurrentneuralnetworks,suchasLong Short TermMemory(LSTM),topredictthecoordinatesoftheguns.Inthisway,itispossible to haveapredictionofarmedpeopleatalltimes.ThemeasurementbetweentheYOLO handgun detectionboundingboxesandtheLSTMpredictionresultedinanIoUof65.23%. When thefirearmdetectionbytheobjectdetectorisinterrupted,theweapon’spositionis generated bytheLSTMmodelsthat,togetherwiththeAPDs,identifythearmedpeople. When theLSTMsdeliveredtheirpredictionstotheAPDs,theNB-APDdemonstratedthe best performance,achievinganaccuracyof80.93%.TheLSTMsallowedtheanalysisof 5,288 recordsofthetestvideothatcouldnotbeanalyzedbeforeduetothelackofknowledge of thegun’sposition.
- Un estudio de competitividad automotriz mexicana: producción de vehículos en el periodo T-MEC(Instituto Tecnológico y de Estudios Superiores de Monterrey, 2024-11-26) Pérez Arredondo, Germán; Carreño Huicochea, Mariel; emimmayorquin; Flores Castro, Mario Adrián; Amoros Espinosa, José Ernesto; Escuela de Graduados en Administración y Dirección de Empresas; Campus Ciudad de México; Cerecedo Hernández, DanielLa presente tesis doctoral aborda la competitividad automotriz en México, desde la perspectiva de la producción de vehículos ligeros, como una de las industrias más importantes para el país, que se ha fortalecido a través del tiempo, particularmente por su ventaja comparativa. Se estudian las variables: Producción de Vehículos Ligeros en México, Producto Interno Bruto para la Fabricación de Vehículos Ligeros, Inversión Extranjera Directa para la Fabricación de Automóviles y Camiones, Consumo Interno de Vehículos Ligeros y Exportaciones de Vehículos Ligeros, considerando el periodo de estudio del 2005 al 2023, de tal manera que mediante pronósticos de series de tiempo se vislumbran los escenarios posibles que México presentará en torno al cumplimiento o no de la hipótesis planteada. Finalmente se exponen conclusiones de los hallazgos encontrados en el desarrollo del modelo, así como una propuesta de elementos de innovación y futuras líneas de investigación que contribuirían a lograr un México más competitivo en los años subsecuentes.