Publication: Realidad aumentada como soporte de asistencia y formación integrada en la Industria 4.0
Loading...
Date
2025-07-28
Authors
Morales Méndez, Ginés
item.page.secondaryauthor
Escuela Internacional de Doctorado
item.page.director
Cerro Velázquez, Francisco del
Publisher
Universidad de Murcia
publication.page.editor
publication.page.department
DOI
item.page.type
info:eu-repo/semantics/doctoralThesis
Description
Abstract
La tesis doctoral tiene como finalidad explorar el papel de la realidad aumentada (RA) como tecnología habilitadora para optimizar la asistencia técnica y la formación en entornos industriales propios de la Industria 4.0. Esta investigación se enmarca en la convergencia entre los procesos de digitalización avanzados, los sistemas ciberfísicos y la transformación del trabajo humano, con el objeto de mejorar la eficiencia operativa, reducir errores y reforzar la seguridad industrial.
El objetivo general del estudio es diseñar una arquitectura basada en RA que se adapte dinámicamente a las habilidades del operario, al tipo de tarea y a las condiciones del entorno, proporcionando apoyo operativo y formativo en tiempo real. A partir de este propósito central, se plantean cinco objetivos específicos: (1) identificar factores clave en la implementación de la RA en la industria, (2) cuantificar su impacto frente a métodos tradicionales, (3) desarrollar una arquitectura adaptativa, (4) validar experimentalmente un prototipo funcional y (5) proponer recomendaciones para su integración efectiva.
La investigación se ha desarrollado en cinco fases. En primer lugar, se llevó a cabo una revisión sistemática y un análisis bibliométrico de 60 estudios relevantes, lo que permitió identificar las principales áreas de aplicación de la RA (como el mantenimiento, la formación o la seguridad) y detectar vacíos en la estandarización y la validación empírica.
En segundo lugar, se realizó un metaanálisis cuantitativo con el fin de sintetizar los efectos de la RA sobre indicadores como la eficiencia, la tasa de errores y la carga cognitiva, donde se evidenció el valor añadido de la RA frente a métodos tradicionales.
La tercera fase consistió en el diseño de una arquitectura adaptativa de RA, capaz de modular el contenido visual según el perfil del operario y el contexto del entorno. El diseño contempló variables como la experiencia del usuario, la complejidad de la tarea y los requisitos de seguridad.
A continuación, se procedió al desarrollo y validación experimental de un prototipo basado en Microsoft HoloLens 2, Unity 3D y Vuforia, en un entorno industrial simulado, los ensayos demostraron mejoras significativas en la precisión, en los tiempos de ejecución, en el mantenimiento predictivo y em la reducción de la carga cognitiva.
Por último, se formularon un conjunto de recomendaciones orientadas a facilitar la adopción de la RA en entornos reales, abarcando aspectos técnicos, formativos, organizativos y normativos.
La tesis demuestra que la RA tiene un impacto significativo en la mejora de procesos industriales, al facilitar la toma de decisiones, incrementar la seguridad y proporcionar formación contextualizada en tiempo real. Los resultados avalan la viabilidad de una arquitectura adaptativa que optimiza la interacción humano-máquina y refuerza la transferencia de conocimiento operativo.
No obstante, el estudio también identifica limitaciones, como las carencias estructurales en la digitalización de la industria española, la volatilidad de los dispositivos aumentados y la escasa madurez tecnológica de algunos sectores. Estas barreras justifican la necesidad de seguir investigando en contextos reales, fomentar la interoperabilidad de sistemas y avanzar hacia modelos más escalables e integrables
The present doctoral thesis aims to explore the role of augmented reality (AR) as an enabling technology to optimise technical assistance and training in industrial environments in Industry 4.0. The present research is situated within the paradigm of convergence between advanced digitalisation processes, cyber-physical systems and the transformation of human work. The objective of this research is to enhance operational efficiency, reduce errors and reinforce industrial safety. The overarching aim of the study is to design an AR-based architecture that is capable of dynamically adapting to operator skills, task type and environmental conditions, thereby providing real-time operational and training support. From this central purpose, five specific objectives are proposed: (1) the identification of key factors in the implementation of AR in industry, (2) the quantification of its impact against traditional methods, (3) the development of an adaptive architecture, (4) the experimental validation of a working prototype and (5) the proposal of recommendations for its effective integration. The research was methodically conducted in five phases. A systematic review and bibliometric analysis of 60 relevant studies was conducted to identify the main areas of application of AR, including maintenance, training and security. The analysis also detected gaps in standardisation and empirical validation. Secondly, a quantitative meta-analysis was conducted to synthesise the effects of AR on indicators such as efficiency, error rate and cognitive load, showing the added value of AR compared to traditional methods. The third phase of the project entailed the conceptualisation of an adaptive AR architecture, which was designed to adapt the visual content according to the operator's profile and the context of the environment. The design incorporated a range of factors, including user experience, task complexity and safety requirements. Subsequently, a prototype was developed and experimentally validated in a simulated industrial environment. This prototype was based on Microsoft HoloLens 2, Unity 3D and Vuforia. The experimental validation process revealed significant improvements in accuracy, execution times, predictive maintenance and reduced cognitive load. A series of recommendations were devised with the objective of encouraging the implementation of AR in actual environments. These recommendations encompass a range of technical, training, organisational and regulatory considerations. The thesis demonstrates that AR has a significant impact on the improvement of industrial processes by facilitating decision-making, increasing safety and providing contextualised training in real time. The results support the feasibility of an adaptive architecture that optimises human-machine interaction and enhances operational knowledge transfer. However, the study also identifies limitations, such as structural shortcomings in the digitisation of Spanish industry, the volatility of augmented devices and the low technological maturity of some sectors. The existence of these barriers underscores the necessity for additional research in authentic contexts, with a view to fostering system interoperability and progressing towards more scalable and integrable models.
The present doctoral thesis aims to explore the role of augmented reality (AR) as an enabling technology to optimise technical assistance and training in industrial environments in Industry 4.0. The present research is situated within the paradigm of convergence between advanced digitalisation processes, cyber-physical systems and the transformation of human work. The objective of this research is to enhance operational efficiency, reduce errors and reinforce industrial safety. The overarching aim of the study is to design an AR-based architecture that is capable of dynamically adapting to operator skills, task type and environmental conditions, thereby providing real-time operational and training support. From this central purpose, five specific objectives are proposed: (1) the identification of key factors in the implementation of AR in industry, (2) the quantification of its impact against traditional methods, (3) the development of an adaptive architecture, (4) the experimental validation of a working prototype and (5) the proposal of recommendations for its effective integration. The research was methodically conducted in five phases. A systematic review and bibliometric analysis of 60 relevant studies was conducted to identify the main areas of application of AR, including maintenance, training and security. The analysis also detected gaps in standardisation and empirical validation. Secondly, a quantitative meta-analysis was conducted to synthesise the effects of AR on indicators such as efficiency, error rate and cognitive load, showing the added value of AR compared to traditional methods. The third phase of the project entailed the conceptualisation of an adaptive AR architecture, which was designed to adapt the visual content according to the operator's profile and the context of the environment. The design incorporated a range of factors, including user experience, task complexity and safety requirements. Subsequently, a prototype was developed and experimentally validated in a simulated industrial environment. This prototype was based on Microsoft HoloLens 2, Unity 3D and Vuforia. The experimental validation process revealed significant improvements in accuracy, execution times, predictive maintenance and reduced cognitive load. A series of recommendations were devised with the objective of encouraging the implementation of AR in actual environments. These recommendations encompass a range of technical, training, organisational and regulatory considerations. The thesis demonstrates that AR has a significant impact on the improvement of industrial processes by facilitating decision-making, increasing safety and providing contextualised training in real time. The results support the feasibility of an adaptive architecture that optimises human-machine interaction and enhances operational knowledge transfer. However, the study also identifies limitations, such as structural shortcomings in the digitisation of Spanish industry, the volatility of augmented devices and the low technological maturity of some sectors. The existence of these barriers underscores the necessity for additional research in authentic contexts, with a view to fostering system interoperability and progressing towards more scalable and integrable models.
publication.page.subject
Citation
item.page.embargo
Collections
Ir a Estadísticas
Este ítem está sujeto a una licencia Creative Commons. CC BY 4.0