Search

Since indoor clothing insulation is a key element in thermal comfort models, the aim of the present study is proposing an approach for predicting it, which could assist the occupants of a building in terms of recommendations regarding their ensemble. For that, a systematic analysis of input variables is exposed, and 13 regression and 12 classification machine learning algorithms were developed and compared. The results are based on data from 3352 questionnaires and 21 input variables from a field study in mixed-mode office buildings in Spain. Outdoor temperature at 6 a.m., indoor air temperature, indoor relative humidity, comfort temperature and gender were the most relevant features for predicting clothing insulation. When comparing machine learning algorithms, decision tree-based algorithms with Boosting techniques achieved the best performance. The proposed model provides an efficient method for forecasting the clothing insulation level and its application would entail optimising thermal comfort and energy efficiency.

Las redes de abastecimiento de agua y saneamiento son infraestructuras críticas que prestan un servicio básico a la sociedad. Estos sistemas, que por tanto deben funcionar correctamente durante su vida útil, muestran un inevitable envejecimiento y degradación, lo que conduce a ineficiencias e impactos en el servicio. Así, una adecuada estrategia de mantenimiento y renovación que garantice la futura sostenibilidad técnica, social, económica y ambiental de estos sistemas se presenta como el reto que las empresas de agua y los municipios deben afrontar en el siglo XXI. Aquí entra en juego la gestión patrimonial de infraestructuras (GPI), que aborda la correcta estimación y planificación de las inversiones necesarias para garantizar su sostenibilidad económica, técnica, social y medioambiental para las futuras generaciones. En particular, esta tesis tiene como objetivo el desarrollo de una metodología GPI eficaz en redes de abastecimiento y saneamiento de aguas, mediante una estimación adecuada del estado general de la red, la programación eficiente de las actividades de renovación y la planificación justificada de las necesidades de inversión que garanticen su sostenibilidad a largo plazo. Las principales propuestas de esta metodología pasan por el desarrollo de un novedoso índice de riesgo para establecer una prioridad de intervención en cada elemento de la red. Para ello se ha empleado un modelo de sumas ponderadas y tenido en cuenta cinco factores asociados al estado de la tubería y a las consecuencias de un posible fallo. También se ha propuesto la utilización de elementos urbanos frente a los hidráulicos como unidades operacionales de intervención, lo que aporta un enfoque más práctico y la posibilidad de cohesionarse mejor con otros posibles proyectos urbanos de mantenimiento (gas, luz o pavimento). Por otro lado, resulta esencial alinear los planes de acción de renovación de la red con las estrategias reales de sustitución de las empresas de agua. En este sentid, Water supply and sewer networks are critical infrastructures that provide a basic service to society. These systems, which must therefore operate correctly during their useful life, show an inevitable aging and degradation, leading to inefficiencies and impacts on the service. Thus, an adequate maintenance and renewal strategy that guarantees the future technical, social, economic and environmental sustainability of these systems is presented as the challenge that water companies and municipalities must face in the 21st century. This is where infrastructure asset management (IAM) comes into play, which addresses the correct estimation and planning of the necessary investments to guarantee their economic, technical, social and environmental sustainability for future generations. In particular, this thesis aims to develop an effective IAM methodology for water supply and sanitation networks, through an adequate estimation of the general condition of the network, the efficient scheduling of renewal activities and the justified planning of investment needs to ensure its long-term sustainability. The main proposals of this methodology include the development of a novel risk index to establish a priority for intervention in each element of the network. For this purpose, a weighted sum model has been used and five factors associated with the condition of the pipeline and the consequences of a possible failure have been taken into account. Also, urban elements rather than hydraulic elements haven been considered as operational units of replacement, which provides a more practical approach and the possibility of better coordination with other possible urban maintenance projects. On the other hand, it is essential to align the network renewal action plans with the actual replacement strategies of the water companies. In this sense, operational criteria of adjacency and synergy of simultaneous replacement, not addressed so far, have been integrated with other criteria of budge

Los puertos son actores fundamentales en las cadenas de suministro de multitud de productos en todo el mundo. A través de las terminales intermodales de contenedores transita un elevado porcentaje de los bienes de consumo y de producción transportados globalmente, por lo que su eficiencia logística debe ser una prioridad dentro de la gestión de la cadena de suministro. Sin embargo, las terminales portuarias son sistemas extremadamente complejos, en los que intervienen multitud de recursos materiales y humanos, de cuya planificación depende el rendimiento de las operaciones de transporte y manejo de la mercancía y, consecuentemente, el nivel de servicio ofrecido a los clientes del puerto. En última instancia, un servicio ineficiente puede llegar a impactar a los consumidores, tanto en términos económicos como de nivel de servicio, afectado por los potenciales retrasos en los plazos de entrega. En esta tesis doctoral se presentan dos acercamientos a la gestión de terminales portuarias, encuadrados en el caso de estudio del Puerto de Sevilla. En concreto, se presenta primero el problema de la planificación de las operaciones logísticas en una terminal intermodal de contenedores. Para abordar este problema, se describe un acercamiento de optimización exacta a la planificación integrada de las operaciones de descarga y carga en los distintos modos de transporte (buques, trenes y camiones) basado en un modelo de programación lineal entero mixto. Este enfoque se considera una propuesta integrada, ya que engloba los problemas de determinación de los tiempos de llegada y salida de las unidades de transporte a la terminal; asignación de muelles de atraque a los buques; secuenciación de la descarga y la carga en los trenes; y asignación y programación de las operaciones de descarga y carga en cada grúa. Este mismo problema se ha resuelto posteriormente empleando un procedimiento heurístico, basado en la combinación de una rutina heurística de construcción de soluciones con tre

Smart retail technologies are making their way into stores at a fast pace, as managers identify them as a great opportunity to save costs and attract clients into their shopping establishments. However, customers’ acceptance of such innovations cannot be taken for granted. Precisely, managers’ concerns about their clients’ perception of in-store innovations are one of the main obstacles for the spread of smart retail technologies. In this article, the results of a survey completed by 2,010 customers concerning the introduction in the Spanish retail sector of a self-checkout system known as Scan and Go are analyzed. This tool allows customers to use their smartphones to scan the desired products, which are automatically added to their tab. In order to finish the purchase and leave the store, clients must simply show a code generated by the app to an employee. The paper analyzes and discusses the results of the survey, mainly the respondents’ usage intention of the tool, focusing on aspects such as determining factors discussed in the literature, consumption habits and socio-economic background, and studying their influence on customer acceptance. Evidence of a positive response from customers to the introduction of the Scan and Go tool in Spanish retail establishments is found, as well as a target customer profile which can serve as a starting market segment for the deployment of the system., Las tecnologías de venta inteligentes están cada vez más presentes en las tiendas físicas, pero los gerentes de muchos puntos de venta siguen teniendo dudas acerca de su aceptación por parte de los consumidores. En este artículo se presentan los resultados de una encuesta administrada a 2,010 clientes acerca de la introducción de un sistema de auto-pago denominado Scan and Go en tiendas españolas y se analiza el impacto de diversos factores y hábitos de consumo en la aceptación de la tecnología. Además, se identifica una respuesta positiva por parte de los usuarios españoles y se determina un perfil de cliente objetivo.

Las redes de abastecimiento de agua y saneamiento son infraestructuras críticas que prestan un servicio básico a la sociedad. Estos sistemas, que por tanto deben funcionar correctamente durante su vida útil, muestran un inevitable envejecimiento y degradación, lo que conduce a ineficiencias e impactos en el servicio. Así, una adecuada estrategia de mantenimiento y renovación que garantice la futura sostenibilidad técnica, social, económica y ambiental de estos sistemas se presenta como el reto que las empresas de agua y los municipios deben afrontar en el siglo XXI. Aquí entra en juego la gestión patrimonial de infraestructuras (GPI), que aborda la correcta estimación y planificación de las inversiones necesarias para garantizar su sostenibilidad económica, técnica, social y medioambiental para las futuras generaciones. En particular, esta tesis tiene como objetivo el desarrollo de una metodología GPI eficaz en redes de abastecimiento y saneamiento de aguas, mediante una estimación adecuada del estado general de la red, la programación eficiente de las actividades de renovación y la planificación justificada de las necesidades de inversión que garanticen su sostenibilidad a largo plazo. Las principales propuestas de esta metodología pasan por el desarrollo de un novedoso índice de riesgo para establecer una prioridad de intervención en cada elemento de la red. Para ello se ha empleado un modelo de sumas ponderadas y tenido en cuenta cinco factores asociados al estado de la tubería y a las consecuencias de un posible fallo. También se ha propuesto la utilización de elementos urbanos frente a los hidráulicos como unidades operacionales de intervención, lo que aporta un enfoque más práctico y la posibilidad de cohesionarse mejor con otros posibles proyectos urbanos de mantenimiento (gas, luz o pavimento). Por otro lado, resulta esencial alinear los planes de acción de renovación de la red con las estrategias reales de sustitución de las empresas de agua. En este sentid, Water supply and sewer networks are critical infrastructures that provide a basic service to society. These systems, which must therefore operate correctly during their useful life, show an inevitable aging and degradation, leading to inefficiencies and impacts on the service. Thus, an adequate maintenance and renewal strategy that guarantees the future technical, social, economic and environmental sustainability of these systems is presented as the challenge that water companies and municipalities must face in the 21st century. This is where infrastructure asset management (IAM) comes into play, which addresses the correct estimation and planning of the necessary investments to guarantee their economic, technical, social and environmental sustainability for future generations. In particular, this thesis aims to develop an effective IAM methodology for water supply and sanitation networks, through an adequate estimation of the general condition of the network, the efficient scheduling of renewal activities and the justified planning of investment needs to ensure its long-term sustainability. The main proposals of this methodology include the development of a novel risk index to establish a priority for intervention in each element of the network. For this purpose, a weighted sum model has been used and five factors associated with the condition of the pipeline and the consequences of a possible failure have been taken into account. Also, urban elements rather than hydraulic elements haven been considered as operational units of replacement, which provides a more practical approach and the possibility of better coordination with other possible urban maintenance projects. On the other hand, it is essential to align the network renewal action plans with the actual replacement strategies of the water companies. In this sense, operational criteria of adjacency and synergy of simultaneous replacement, not addressed so far, have been integrated with other criteria of budge

Smart retail technologies are making their way into stores at a fast pace, as managers identify them as a great opportunity to save costs and attract clients into their shopping establishments. However, customers’ acceptance of such innovations cannot be taken for granted. Precisely, managers’ concerns about their clients’ perception of in-store innovations are one of the main obstacles for the spread of smart retail technologies. In this article, the results of a survey completed by 2,010 customers concerning the introduction in the Spanish retail sector of a self-checkout system known as Scan and Go are analyzed. This tool allows customers to use their smartphones to scan the desired products, which are automatically added to their tab. In order to finish the purchase and leave the store, clients must simply show a code generated by the app to an employee. The paper analyzes and discusses the results of the survey, mainly the respondents’ usage intention of the tool, focusing on aspects such as determining factors discussed in the literature, consumption habits and socio-economic background, and studying their influence on customer acceptance. Evidence of a positive response from customers to the introduction of the Scan and Go tool in Spanish retail establishments is found, as well as a target customer profile which can serve as a starting market segment for the deployment of the system., Las tecnologías de venta inteligentes están cada vez más presentes en las tiendas físicas, pero los gerentes de muchos puntos de venta siguen teniendo dudas acerca de su aceptación por parte de los consumidores. En este artículo se presentan los resultados de una encuesta administrada a 2,010 clientes acerca de la introducción de un sistema de auto-pago denominado Scan and Go en tiendas españolas y se analiza el impacto de diversos factores y hábitos de consumo en la aceptación de la tecnología. Además, se identifica una respuesta positiva por parte de los usuarios españoles y se determina un perfil de cliente objetivo.

The unexpected failure of pipes is a problem that is hitting the water networks of many cities around the world. Nowadays, many proposals based on the use of machine learning techniques are emerging to combat this problem. However, most studies focus their efforts on predicting failures in short time periods, usually a year, while longer time period predictions would be more valuable to address strategic decisions.In this study, the use of multi-label classification techniques is proposed to simultaneously predict pipe failures in water supply systems for multiple years. For this purpose, three models (discriminant analysis, logistic regression and random forest) and different prediction time periods (one, two and three years) have been analysed. As multi-label data require specific quality metrics and sampling techniques, part of this work is dedicated to their exploration and discussion.The models are evaluated on a real-world seven-year database, achieving successful results. An insightful analysis of the use of the methodology shows how the percentage of avoided pipe failures increases over time. In fact, it is demonstrated that 30.2%, 51.4% and 54.0% of the pipe failures of three consecutive years are avoided according to data from a real network.

Los puertos son actores fundamentales en las cadenas de suministro de multitud de productos en todo el mundo. A través de las terminales intermodales de contenedores transita un elevado porcentaje de los bienes de consumo y de producción transportados globalmente, por lo que su eficiencia logística debe ser una prioridad dentro de la gestión de la cadena de suministro. Sin embargo, las terminales portuarias son sistemas extremadamente complejos, en los que intervienen multitud de recursos materiales y humanos, de cuya planificación depende el rendimiento de las operaciones de transporte y manejo de la mercancía y, consecuentemente, el nivel de servicio ofrecido a los clientes del puerto. En última instancia, un servicio ineficiente puede llegar a impactar a los consumidores, tanto en términos económicos como de nivel de servicio, afectado por los potenciales retrasos en los plazos de entrega. En esta tesis doctoral se presentan dos acercamientos a la gestión de terminales portuarias, encuadrados en el caso de estudio del Puerto de Sevilla. En concreto, se presenta primero el problema de la planificación de las operaciones logísticas en una terminal intermodal de contenedores. Para abordar este problema, se describe un acercamiento de optimización exacta a la planificación integrada de las operaciones de descarga y carga en los distintos modos de transporte (buques, trenes y camiones) basado en un modelo de programación lineal entero mixto. Este enfoque se considera una propuesta integrada, ya que engloba los problemas de determinación de los tiempos de llegada y salida de las unidades de transporte a la terminal; asignación de muelles de atraque a los buques; secuenciación de la descarga y la carga en los trenes; y asignación y programación de las operaciones de descarga y carga en cada grúa. Este mismo problema se ha resuelto posteriormente empleando un procedimiento heurístico, basado en la combinación de una rutina heurística de construcción de soluciones con tre

Los puertos son actores fundamentales en las cadenas de suministro de multitud de productos en todo el mundo. A través de las terminales intermodales de contenedores transita un elevado porcentaje de los bienes de consumo y de producción transportados globalmente, por lo que su eficiencia logística debe ser una prioridad dentro de la gestión de la cadena de suministro. Sin embargo, las terminales portuarias son sistemas extremadamente complejos, en los que intervienen multitud de recursos materiales y humanos, de cuya planificación depende el rendimiento de las operaciones de transporte y manejo de la mercancía y, consecuentemente, el nivel de servicio ofrecido a los clientes del puerto. En última instancia, un servicio ineficiente puede llegar a impactar a los consumidores, tanto en términos económicos como de nivel de servicio, afectado por los potenciales retrasos en los plazos de entrega. En esta tesis doctoral se presentan dos acercamientos a la gestión de terminales portuarias, encuadrados en el caso de estudio del Puerto de Sevilla. En concreto, se presenta primero el problema de la planificación de las operaciones logísticas en una terminal intermodal de contenedores. Para abordar este problema, se describe un acercamiento de optimización exacta a la planificación integrada de las operaciones de descarga y carga en los distintos modos de transporte (buques, trenes y camiones) basado en un modelo de programación lineal entero mixto. Este enfoque se considera una propuesta integrada, ya que engloba los problemas de determinación de los tiempos de llegada y salida de las unidades de transporte a la terminal; asignación de muelles de atraque a los buques; secuenciación de la descarga y la carga en los trenes; y asignación y programación de las operaciones de descarga y carga en cada grúa. Este mismo problema se ha resuelto posteriormente empleando un procedimiento heurístico, basado en la combinación de una rutina heurística de construcción de soluciones con tre

Las redes de abastecimiento de agua y saneamiento son infraestructuras críticas que prestan un servicio básico a la sociedad. Estos sistemas, que por tanto deben funcionar correctamente durante su vida útil, muestran un inevitable envejecimiento y degradación, lo que conduce a ineficiencias e impactos en el servicio. Así, una adecuada estrategia de mantenimiento y renovación que garantice la futura sostenibilidad técnica, social, económica y ambiental de estos sistemas se presenta como el reto que las empresas de agua y los municipios deben afrontar en el siglo XXI. Aquí entra en juego la gestión patrimonial de infraestructuras (GPI), que aborda la correcta estimación y planificación de las inversiones necesarias para garantizar su sostenibilidad económica, técnica, social y medioambiental para las futuras generaciones. En particular, esta tesis tiene como objetivo el desarrollo de una metodología GPI eficaz en redes de abastecimiento y saneamiento de aguas, mediante una estimación adecuada del estado general de la red, la programación eficiente de las actividades de renovación y la planificación justificada de las necesidades de inversión que garanticen su sostenibilidad a largo plazo. Las principales propuestas de esta metodología pasan por el desarrollo de un novedoso índice de riesgo para establecer una prioridad de intervención en cada elemento de la red. Para ello se ha empleado un modelo de sumas ponderadas y tenido en cuenta cinco factores asociados al estado de la tubería y a las consecuencias de un posible fallo. También se ha propuesto la utilización de elementos urbanos frente a los hidráulicos como unidades operacionales de intervención, lo que aporta un enfoque más práctico y la posibilidad de cohesionarse mejor con otros posibles proyectos urbanos de mantenimiento (gas, luz o pavimento). Por otro lado, resulta esencial alinear los planes de acción de renovación de la red con las estrategias reales de sustitución de las empresas de agua. En este sentid, Water supply and sewer networks are critical infrastructures that provide a basic service to society. These systems, which must therefore operate correctly during their useful life, show an inevitable aging and degradation, leading to inefficiencies and impacts on the service. Thus, an adequate maintenance and renewal strategy that guarantees the future technical, social, economic and environmental sustainability of these systems is presented as the challenge that water companies and municipalities must face in the 21st century. This is where infrastructure asset management (IAM) comes into play, which addresses the correct estimation and planning of the necessary investments to guarantee their economic, technical, social and environmental sustainability for future generations. In particular, this thesis aims to develop an effective IAM methodology for water supply and sanitation networks, through an adequate estimation of the general condition of the network, the efficient scheduling of renewal activities and the justified planning of investment needs to ensure its long-term sustainability. The main proposals of this methodology include the development of a novel risk index to establish a priority for intervention in each element of the network. For this purpose, a weighted sum model has been used and five factors associated with the condition of the pipeline and the consequences of a possible failure have been taken into account. Also, urban elements rather than hydraulic elements haven been considered as operational units of replacement, which provides a more practical approach and the possibility of better coordination with other possible urban maintenance projects. On the other hand, it is essential to align the network renewal action plans with the actual replacement strategies of the water companies. In this sense, operational criteria of adjacency and synergy of simultaneous replacement, not addressed so far, have been integrated with other criteria of budge

This PhD thesis addresses the problem of the appearance of unexpected pipe failures in water distribution networks. Specifically, it seeks to predict such failures using machine learning based techniques. An in depth literature review on the subject informs that although there are studies that have tested certain machine learning techniques for the aforementioned purpose, this is a novel issue that has not been fully explored yet. Consequently, this work proposes several machine learning models, some of which have not been applied to this problem before and analyses the most significant aspects of data processing and evaluation of the results. The nature and characteristics of the data are key points on the design of a machine learning system. For the development of this thesis, the company that manages the water distribution network of Seville (Spain) called EMASESA has provided an extensive database. Concretely, the database consists of a seven year pipe failure history, from 2012 to 2018, and includes various factors related to each of the pipes that compose the more than 3800 kilometres network. The first strategy has been to forecast pipe failures one year in advance, since companies generally decide their maintenance and replacement plans annually. Therefore, and according to the characteristics of the problem and the available data, the following machine learning techniques are proposed: discriminant analysis, logistic regression, support vector machines, random forests, artificial neural networks and evolutionary fuzzy logic. All these models can work as classifiers, being the main part of a supervised classification machine learning system. In this case, the output of the system is defined as a binary variable that takes the value 1 when a pipe fails in the period of study, and 0 otherwise. Secondly, the initial focus of this thesis was extended to multi label classification, which allows predicting more than one output variable at the same time. The aim of, En esta Tesis se aborda el problema de la aparición de roturas o fallos inesperados en las tuberías que componen las redes de distribución de agua. Concretamente, se busca predecir dichas roturas utilizando técnicas basadas en el aprendizaje automático, del inglés machine learning. Tras un análisis exhaustivo de la literatura existente sobre el tema, se detecta que, aunque ya existen estudios que proponen ciertas técnicas de machine learning para el propósito anteriormente descrito, es una temática reciente que aún no ha sido desarrollada en su totalidad. Por ello, este trabajo propone distintos modelos de machine learning, algunos de los cuales no han sido aplicados a la problemática de estudio hasta la fecha, y analiza los aspectos más significativos del procesamiento de los datos y de la evaluación de los resultados. En el desarrollo de un sistema de machine learning tiene especial importancia la forma y características de los datos a utilizar. En este trabajo, se dispone de una extensa base de datos de la red de abastecimiento de agua de Sevilla, la cual ha sido cedida por la Empresa Metropolitana de Abastecimiento y Saneamiento de Aguas de Sevilla (EMASESA), compañía que gestiona dicha red. La base de datos consta de un histórico de roturas de siete años, de 2012 a 2018 inclusive, e incluye diversas variables relacionadas con cada una de las tuberías que forman sus más de 3800 kilómetros de red. Como primer enfoque, se decide explorar la predicción de fallos en las tuberías con un horizonte temporal de un año, dado que las compañías generalmente planifican las tareas de mantenimiento y reposición de la red a un año vista. Por ello, y de acuerdo a las características del problema y a los datos disponibles, se proponen las siguientes técnicas de machine learning: el análisis discriminante, la regresión logística, las máquinas de vector soporte, los bosques aleatorios, las redes neuronales y la lógica difusa evolutiva. Todas estas técnicas tienen la capacidad de a

This PhD thesis addresses the problem of the appearance of unexpected pipe failures in water distribution networks. Specifically, it seeks to predict such failures using machine learning based techniques. An in depth literature review on the subject informs that although there are studies that have tested certain machine learning techniques for the aforementioned purpose, this is a novel issue that has not been fully explored yet. Consequently, this work proposes several machine learning models, some of which have not been applied to this problem before and analyses the most significant aspects of data processing and evaluation of the results. The nature and characteristics of the data are key points on the design of a machine learning system. For the development of this thesis, the company that manages the water distribution network of Seville (Spain) called EMASESA has provided an extensive database. Concretely, the database consists of a seven year pipe failure history, from 2012 to 2018, and includes various factors related to each of the pipes that compose the more than 3800 kilometres network. The first strategy has been to forecast pipe failures one year in advance, since companies generally decide their maintenance and replacement plans annually. Therefore, and according to the characteristics of the problem and the available data, the following machine learning techniques are proposed: discriminant analysis, logistic regression, support vector machines, random forests, artificial neural networks and evolutionary fuzzy logic. All these models can work as classifiers, being the main part of a supervised classification machine learning system. In this case, the output of the system is defined as a binary variable that takes the value 1 when a pipe fails in the period of study, and 0 otherwise. Secondly, the initial focus of this thesis was extended to multi label classification, which allows predicting more than one output variable at the same time. The aim of, En esta Tesis se aborda el problema de la aparición de roturas o fallos inesperados en las tuberías que componen las redes de distribución de agua. Concretamente, se busca predecir dichas roturas utilizando técnicas basadas en el aprendizaje automático, del inglés machine learning. Tras un análisis exhaustivo de la literatura existente sobre el tema, se detecta que, aunque ya existen estudios que proponen ciertas técnicas de machine learning para el propósito anteriormente descrito, es una temática reciente que aún no ha sido desarrollada en su totalidad. Por ello, este trabajo propone distintos modelos de machine learning, algunos de los cuales no han sido aplicados a la problemática de estudio hasta la fecha, y analiza los aspectos más significativos del procesamiento de los datos y de la evaluación de los resultados. En el desarrollo de un sistema de machine learning tiene especial importancia la forma y características de los datos a utilizar. En este trabajo, se dispone de una extensa base de datos de la red de abastecimiento de agua de Sevilla, la cual ha sido cedida por la Empresa Metropolitana de Abastecimiento y Saneamiento de Aguas de Sevilla (EMASESA), compañía que gestiona dicha red. La base de datos consta de un histórico de roturas de siete años, de 2012 a 2018 inclusive, e incluye diversas variables relacionadas con cada una de las tuberías que forman sus más de 3800 kilómetros de red. Como primer enfoque, se decide explorar la predicción de fallos en las tuberías con un horizonte temporal de un año, dado que las compañías generalmente planifican las tareas de mantenimiento y reposición de la red a un año vista. Por ello, y de acuerdo a las características del problema y a los datos disponibles, se proponen las siguientes técnicas de machine learning: el análisis discriminante, la regresión logística, las máquinas de vector soporte, los bosques aleatorios, las redes neuronales y la lógica difusa evolutiva. Todas estas técnicas tienen la capacidad de a

In the current paradigm, where rapid evolution of the retail industry is seen, especially in textile companies, general product assignment in any distribution and supply chain, consisting of a main warehouse and several locations, can mean an important challenge. The replenishing process in shops centers all its attention on dynamically optimizing the inventory of shops attending to a wide range of products, starting from a sales prevision of previous periods. In this context, the objective of this work is centered on optimizing replenishing work in shops for non-perishable goods which lose value as time passes.

The rapid rise of electronic commerce has entailed an increase in logistic complexity, with last-mile logistics being the most critical element in deliveries. Since users prefer goods to be delivered at home, one of the biggest challenges faced by e-commerce is to reduce the number of incidents that occur in the delivery of goods to the homes of customers. In many cases, these deliveries cannot take place because recipients are not at the agreed delivery point, leading to a decrease in the quality of service and an increase in distribution costs. Furthermore, sometimes the delivery policies are not in tune with the customers’ expectations. This work presents a new perspective of the last-mile logistics in the context of multichannel retail, asking customers to provide several delivery locations (at home, at work, at a familiar home, in a shop, in a locker, etc.) associated with different time windows. In addition, the customer could state their preferences about these locations. This work formulates the problem and develops different approaches to solve it. A benchmark is proposed to analyze the performance and limitations. The results reveal that a distribution policy with several locations can improve the efficiency of electronic commerce by reducing delivery costs. The findings of this study have several implications for distribution companies.

In the current paradigm for the retail industry, which is experiencing a rapid evolution, especially in textile companies, the generic problem of product allocation in a distribution and supply chain consisting of one main warehouse and several locations, belonging to different sales channels, is a challenge. The omnichannel replenishment process focuses on dynamically optimizing a shop or intermediate warehouse inventory for a wide range of products based on a forecast of sales, in order to fulfill the demand of all of the channels considered. In this context, the aims of this work were (a) to optimize inventory replenishment for multiple channels and products that are not perishable but devalue over time, and (b) to implement a methodology that combines the benefits of the Particle Swarm Optimization metaheuristic and Simulated Annealing. This study was carried out for different sales periods, channels and product configurations by performing a sensitivity analysis between the way new solutions are updated and the degree of intensification used in local search.

Water supply and sewer networks are critical infrastructures that provide a basic service to society. However, these systems constantly age and degrade over time. In addition, since network infrastructures are so extensive in length, they require a significant investment in maintenance tasks. Hence, within the context of infrastructure asset management (IAM), accurately defining the most efficient investment planning possible is essential to ensure their long-term sustainability. This paper presents an original five-step comprehensive framework to successfully implement an infrastructure asset management strategy and plan long-term investments. Moreover, this methodology integrates innovative and relevant operational and convenience factors that, while provide the problem both with realism and practicality, have not been addressed so far. To illustrate the usefulness and applicability of this methodology, the case study of a large water company in Spain is presented.

In the current paradigm, where rapid evolution of the retail industry is seen, especially in textile companies, general product assignment in any distribution and supply chain, consisting of a main warehouse and several locations, can mean an important challenge. The replenishing process in shops centers all its attention on dynamically optimizing the inventory of shops attending to a wide range of products, starting from a sales prevision of previous periods. In this context, the objective of this work is centered on optimizing replenishing work in shops for non-perishable goods which lose value as time passes.

The rapid rise of electronic commerce has entailed an increase in logistic complexity, with last-mile logistics being the most critical element in deliveries. Since users prefer goods to be delivered at home, one of the biggest challenges faced by e-commerce is to reduce the number of incidents that occur in the delivery of goods to the homes of customers. In many cases, these deliveries cannot take place because recipients are not at the agreed delivery point, leading to a decrease in the quality of service and an increase in distribution costs. Furthermore, sometimes the delivery policies are not in tune with the customers’ expectations. This work presents a new perspective of the last-mile logistics in the context of multichannel retail, asking customers to provide several delivery locations (at home, at work, at a familiar home, in a shop, in a locker, etc.) associated with different time windows. In addition, the customer could state their preferences about these locations. This work formulates the problem and develops different approaches to solve it. A benchmark is proposed to analyze the performance and limitations. The results reveal that a distribution policy with several locations can improve the efficiency of electronic commerce by reducing delivery costs. The findings of this study have several implications for distribution companies.

In the current paradigm for the retail industry, which is experiencing a rapid evolution, especially in textile companies, the generic problem of product allocation in a distribution and supply chain consisting of one main warehouse and several locations, belonging to different sales channels, is a challenge. The omnichannel replenishment process focuses on dynamically optimizing a shop or intermediate warehouse inventory for a wide range of products based on a forecast of sales, in order to fulfill the demand of all of the channels considered. In this context, the aims of this work were (a) to optimize inventory replenishment for multiple channels and products that are not perishable but devalue over time, and (b) to implement a methodology that combines the benefits of the Particle Swarm Optimization metaheuristic and Simulated Annealing. This study was carried out for different sales periods, channels and product configurations by performing a sensitivity analysis between the way new solutions are updated and the degree of intensification used in local search.

Water supply and sewer networks are critical infrastructures that provide a basic service to society. However, these systems constantly age and degrade over time. In addition, since network infrastructures are so extensive in length, they require a significant investment in maintenance tasks. Hence, within the context of infrastructure asset management (IAM), accurately defining the most efficient investment planning possible is essential to ensure their long-term sustainability. This paper presents an original five-step comprehensive framework to successfully implement an infrastructure asset management strategy and plan long-term investments. Moreover, this methodology integrates innovative and relevant operational and convenience factors that, while provide the problem both with realism and practicality, have not been addressed so far. To illustrate the usefulness and applicability of this methodology, the case study of a large water company in Spain is presented.

This PhD thesis addresses the problem of the appearance of unexpected pipe failures in water distribution networks. Specifically, it seeks to predict such failures using machine learning based techniques. An in depth literature review on the subject informs that although there are studies that have tested certain machine learning techniques for the aforementioned purpose, this is a novel issue that has not been fully explored yet. Consequently, this work proposes several machine learning models, some of which have not been applied to this problem before and analyses the most significant aspects of data processing and evaluation of the results. The nature and characteristics of the data are key points on the design of a machine learning system. For the development of this thesis, the company that manages the water distribution network of Seville (Spain) called EMASESA has provided an extensive database. Concretely, the database consists of a seven year pipe failure history, from 2012 to 2018, and includes various factors related to each of the pipes that compose the more than 3800 kilometres network. The first strategy has been to forecast pipe failures one year in advance, since companies generally decide their maintenance and replacement plans annually. Therefore, and according to the characteristics of the problem and the available data, the following machine learning techniques are proposed: discriminant analysis, logistic regression, support vector machines, random forests, artificial neural networks and evolutionary fuzzy logic. All these models can work as classifiers, being the main part of a supervised classification machine learning system. In this case, the output of the system is defined as a binary variable that takes the value 1 when a pipe fails in the period of study, and 0 otherwise. Secondly, the initial focus of this thesis was extended to multi label classification, which allows predicting more than one output variable at the same time. The aim of, En esta Tesis se aborda el problema de la aparición de roturas o fallos inesperados en las tuberías que componen las redes de distribución de agua. Concretamente, se busca predecir dichas roturas utilizando técnicas basadas en el aprendizaje automático, del inglés machine learning. Tras un análisis exhaustivo de la literatura existente sobre el tema, se detecta que, aunque ya existen estudios que proponen ciertas técnicas de machine learning para el propósito anteriormente descrito, es una temática reciente que aún no ha sido desarrollada en su totalidad. Por ello, este trabajo propone distintos modelos de machine learning, algunos de los cuales no han sido aplicados a la problemática de estudio hasta la fecha, y analiza los aspectos más significativos del procesamiento de los datos y de la evaluación de los resultados. En el desarrollo de un sistema de machine learning tiene especial importancia la forma y características de los datos a utilizar. En este trabajo, se dispone de una extensa base de datos de la red de abastecimiento de agua de Sevilla, la cual ha sido cedida por la Empresa Metropolitana de Abastecimiento y Saneamiento de Aguas de Sevilla (EMASESA), compañía que gestiona dicha red. La base de datos consta de un histórico de roturas de siete años, de 2012 a 2018 inclusive, e incluye diversas variables relacionadas con cada una de las tuberías que forman sus más de 3800 kilómetros de red. Como primer enfoque, se decide explorar la predicción de fallos en las tuberías con un horizonte temporal de un año, dado que las compañías generalmente planifican las tareas de mantenimiento y reposición de la red a un año vista. Por ello, y de acuerdo a las características del problema y a los datos disponibles, se proponen las siguientes técnicas de machine learning: el análisis discriminante, la regresión logística, las máquinas de vector soporte, los bosques aleatorios, las redes neuronales y la lógica difusa evolutiva. Todas estas técnicas tienen la capacidad de a

In Spain, the water supply service is a municipal responsibility and in general is a sector without competitors. For this reason, an efficiency analysis attains greater significance. This study uses Data Envelopment Analysis (DEA) to estimate the efficiency of different urban water utilities. An extensive data search, where several variables such as the capital expenditure, the cost of material, or the labor have been recorded, has allowed evaluating the relative efficiency of the most important Spanish water distribution networks in using their resources. Furthermore, their sustainable efficiency has also been evaluated by including a variable representing the percentage of water losses. Results reveal the weaknesses of inefficient utilities and help to detect potential aspects that these companies should improve. For instance, there is an evident incorrect management of the costs of material by many urban water utilities, which does not happen with the labor. Additionally, the most efficient water utilities regarding the sustainable efficiency help to discover target percentages of water losses for the inefficient ones.

The multiple processes taking place on a daily basis at an intermodal container terminal are often considered individually, given the complexity of their joint consideration. Nevertheless, the integrated planning and scheduling of operations in an intermodal terminal, including the arrivals and departures of trains and vessels, is a very relevant topic for terminal managers, which can benefit from the application of Operations Research (OR) techniques to obtain near-optimal solutions without excessive computational cost. Applying the functional integration technique, we present here a mathematical model for this terminal planning process, and solve it using heuristic procedures, given its complexity and size. Details on the benchmark comparison of a genetic algorithm, a simulated annealing routine and a tabu search are provided for different problem instances.

In Spain, the water supply service is a municipal responsibility and in general is a sector without competitors. For this reason, an efficiency analysis attains greater significance. This study uses Data Envelopment Analysis (DEA) to estimate the efficiency of different urban water utilities. An extensive data search, where several variables such as the capital expenditure, the cost of material, or the labor have been recorded, has allowed evaluating the relative efficiency of the most important Spanish water distribution networks in using their resources. Furthermore, their sustainable efficiency has also been evaluated by including a variable representing the percentage of water losses. Results reveal the weaknesses of inefficient utilities and help to detect potential aspects that these companies should improve. For instance, there is an evident incorrect management of the costs of material by many urban water utilities, which does not happen with the labor. Additionally, the most efficient water utilities regarding the sustainable efficiency help to discover target percentages of water losses for the inefficient ones.

The multiple processes taking place on a daily basis at an intermodal container terminal are often considered individually, given the complexity of their joint consideration. Nevertheless, the integrated planning and scheduling of operations in an intermodal terminal, including the arrivals and departures of trains and vessels, is a very relevant topic for terminal managers, which can benefit from the application of Operations Research (OR) techniques to obtain near-optimal solutions without excessive computational cost. Applying the functional integration technique, we present here a mathematical model for this terminal planning process, and solve it using heuristic procedures, given its complexity and size. Details on the benchmark comparison of a genetic algorithm, a simulated annealing routine and a tabu search are provided for different problem instances.

Water supply and sewer networks are critical infrastructures that provide a basic service to society. However, these systems constantly age and degrade over time. In addition, since network infrastructures are so extensive in length, they require a significant investment in maintenance tasks. Hence, within the context of infrastructure asset management (IAM), accurately defining the most efficient investment planning possible is essential to ensure their long-term sustainability. This paper presents an original five-step comprehensive framework to successfully implement an infrastructure asset management strategy and plan long-term investments. Moreover, this methodology integrates innovative and relevant operational and convenience factors that, while provide the problem both with realism and practicality, have not been addressed so far. To illustrate the usefulness and applicability of this methodology, the case study of a large water company in Spain is presented.

La gestión de los activos y su renovación generalmente se planifican a nivel anual y operativo. Sin embargo, a nivel estratégico y táctico, las empresas de infraestructuras buscan modelos de gestión que permitan analizar sus activos, su evolución y la detección de aquéllos que deben ser renovados e incorporados antes en las planificaciones a medio y largo plazo. Además, es necesario definir el esfuerzo financiero y su aplicación temporal. Dicho esfuerzo debe permitir que la cartera de activos que conforman la red hídrica esté alineada con los objetivos de gestión de activos de la organización. El seguimiento de los objetivos, entre ellos el mantenimiento sostenible de la infraestructura de red, se puede desarrollar en base a indicadores de gestión que faciliten su renovación. Los indicadores permiten analizar el estado o fortaleza de la organización y constituyen un factor de oportunidad para mejorar la eficiencia en la administración de los activos cuando se combinan las estrategias financieras y las no financieras.

In the current paradigm, where rapid evolution of the retail industry is seen, especially in textile companies, general product assignment in any distribution and supply chain, consisting of a main warehouse and several locations, can mean an important challenge. The replenishing process in shops centers all its attention on dynamically optimizing the inventory of shops attending to a wide range of products, starting from a sales prevision of previous periods. In this context, the objective of this work is centered on optimizing replenishing work in shops for non-perishable goods which lose value as time passes.

In the current paradigm for the retail industry, which is experiencing a rapid evolution, especially in textile companies, the generic problem of product allocation in a distribution and supply chain consisting of one main warehouse and several locations, belonging to different sales channels, is a challenge. The omnichannel replenishment process focuses on dynamically optimizing a shop or intermediate warehouse inventory for a wide range of products based on a forecast of sales, in order to fulfill the demand of all of the channels considered. In this context, the aims of this work were (a) to optimize inventory replenishment for multiple channels and products that are not perishable but devalue over time, and (b) to implement a methodology that combines the benefits of the Particle Swarm Optimization metaheuristic and Simulated Annealing. This study was carried out for different sales periods, channels and product configurations by performing a sensitivity analysis between the way new solutions are updated and the degree of intensification used in local search.

The rapid rise of electronic commerce has entailed an increase in logistic complexity, with last-mile logistics being the most critical element in deliveries. Since users prefer goods to be delivered at home, one of the biggest challenges faced by e-commerce is to reduce the number of incidents that occur in the delivery of goods to the homes of customers. In many cases, these deliveries cannot take place because recipients are not at the agreed delivery point, leading to a decrease in the quality of service and an increase in distribution costs. Furthermore, sometimes the delivery policies are not in tune with the customers’ expectations. This work presents a new perspective of the last-mile logistics in the context of multichannel retail, asking customers to provide several delivery locations (at home, at work, at a familiar home, in a shop, in a locker, etc.) associated with different time windows. In addition, the customer could state their preferences about these locations. This work formulates the problem and develops different approaches to solve it. A benchmark is proposed to analyze the performance and limitations. The results reveal that a distribution policy with several locations can improve the efficiency of electronic commerce by reducing delivery costs. The findings of this study have several implications for distribution companies.

The multiple processes taking place on a daily basis at an intermodal container terminal are often considered individually, given the complexity of their joint consideration. Nevertheless, the integrated planning and scheduling of operations in an intermodal terminal, including the arrivals and departures of trains and vessels, is a very relevant topic for terminal managers, which can benefit from the application of Operations Research (OR) techniques to obtain near-optimal solutions without excessive computational cost. Applying the functional integration technique, we present here a mathematical model for this terminal planning process, and solve it using heuristic procedures, given its complexity and size. Details on the benchmark comparison of a genetic algorithm, a simulated annealing routine and a tabu search are provided for different problem instances.

La gestión de los activos y su renovación generalmente se planifican a nivel anual y operativo. Sin embargo, a nivel estratégico y táctico, las empresas de infraestructuras buscan modelos de gestión que permitan analizar sus activos, su evolución y la detección de aquéllos que deben ser renovados e incorporados antes en las planificaciones a medio y largo plazo. Además, es necesario definir el esfuerzo financiero y su aplicación temporal. Dicho esfuerzo debe permitir que la cartera de activos que conforman la red hídrica esté alineada con los objetivos de gestión de activos de la organización. El seguimiento de los objetivos, entre ellos el mantenimiento sostenible de la infraestructura de red, se puede desarrollar en base a indicadores de gestión que faciliten su renovación. Los indicadores permiten analizar el estado o fortaleza de la organización y constituyen un factor de oportunidad para mejorar la eficiencia en la administración de los activos cuando se combinan las estrategias financieras y las no financieras.

In the current paradigm, where rapid evolution of the retail industry is seen, especially in textile companies, general product assignment in any distribution and supply chain, consisting of a main warehouse and several locations, can mean an important challenge. The replenishing process in shops centers all its attention on dynamically optimizing the inventory of shops attending to a wide range of products, starting from a sales prevision of previous periods. In this context, the objective of this work is centered on optimizing replenishing work in shops for non-perishable goods which lose value as time passes.

The rapid rise of electronic commerce has entailed an increase in logistic complexity, with last-mile logistics being the most critical element in deliveries. Since users prefer goods to be delivered at home, one of the biggest challenges faced by e-commerce is to reduce the number of incidents that occur in the delivery of goods to the homes of customers. In many cases, these deliveries cannot take place because recipients are not at the agreed delivery point, leading to a decrease in the quality of service and an increase in distribution costs. Furthermore, sometimes the delivery policies are not in tune with the customers’ expectations. This work presents a new perspective of the last-mile logistics in the context of multichannel retail, asking customers to provide several delivery locations (at home, at work, at a familiar home, in a shop, in a locker, etc.) associated with different time windows. In addition, the customer could state their preferences about these locations. This work formulates the problem and develops different approaches to solve it. A benchmark is proposed to analyze the performance and limitations. The results reveal that a distribution policy with several locations can improve the efficiency of electronic commerce by reducing delivery costs. The findings of this study have several implications for distribution companies.

In the current paradigm for the retail industry, which is experiencing a rapid evolution, especially in textile companies, the generic problem of product allocation in a distribution and supply chain consisting of one main warehouse and several locations, belonging to different sales channels, is a challenge. The omnichannel replenishment process focuses on dynamically optimizing a shop or intermediate warehouse inventory for a wide range of products based on a forecast of sales, in order to fulfill the demand of all of the channels considered. In this context, the aims of this work were (a) to optimize inventory replenishment for multiple channels and products that are not perishable but devalue over time, and (b) to implement a methodology that combines the benefits of the Particle Swarm Optimization metaheuristic and Simulated Annealing. This study was carried out for different sales periods, channels and product configurations by performing a sensitivity analysis between the way new solutions are updated and the degree of intensification used in local search.

Water supply and sewer networks are critical infrastructures that provide a basic service to society. However, these systems constantly age and degrade over time. In addition, since network infrastructures are so extensive in length, they require a significant investment in maintenance tasks. Hence, within the context of infrastructure asset management (IAM), accurately defining the most efficient investment planning possible is essential to ensure their long-term sustainability. This paper presents an original five-step comprehensive framework to successfully implement an infrastructure asset management strategy and plan long-term investments. Moreover, this methodology integrates innovative and relevant operational and convenience factors that, while provide the problem both with realism and practicality, have not been addressed so far. To illustrate the usefulness and applicability of this methodology, the case study of a large water company in Spain is presented.

In Spain, the water supply service is a municipal responsibility and in general is a sector without competitors. For this reason, an efficiency analysis attains greater significance. This study uses Data Envelopment Analysis (DEA) to estimate the efficiency of different urban water utilities. An extensive data search, where several variables such as the capital expenditure, the cost of material, or the labor have been recorded, has allowed evaluating the relative efficiency of the most important Spanish water distribution networks in using their resources. Furthermore, their sustainable efficiency has also been evaluated by including a variable representing the percentage of water losses. Results reveal the weaknesses of inefficient utilities and help to detect potential aspects that these companies should improve. For instance, there is an evident incorrect management of the costs of material by many urban water utilities, which does not happen with the labor. Additionally, the most efficient water utilities regarding the sustainable efficiency help to discover target percentages of water losses for the inefficient ones.

This PhD thesis addresses the problem of the appearance of unexpected pipe failures in water distribution networks. Specifically, it seeks to predict such failures using machine learning based techniques. An in depth literature review on the subject informs that although there are studies that have tested certain machine learning techniques for the aforementioned purpose, this is a novel issue that has not been fully explored yet. Consequently, this work proposes several machine learning models, some of which have not been applied to this problem before and analyses the most significant aspects of data processing and evaluation of the results. The nature and characteristics of the data are key points on the design of a machine learning system. For the development of this thesis, the company that manages the water distribution network of Seville (Spain) called EMASESA has provided an extensive database. Concretely, the database consists of a seven year pipe failure history, from 2012 to 2018, and includes various factors related to each of the pipes that compose the more than 3800 kilometres network. The first strategy has been to forecast pipe failures one year in advance, since companies generally decide their maintenance and replacement plans annually. Therefore, and according to the characteristics of the problem and the available data, the following machine learning techniques are proposed: discriminant analysis, logistic regression, support vector machines, random forests, artificial neural networks and evolutionary fuzzy logic. All these models can work as classifiers, being the main part of a supervised classification machine learning system. In this case, the output of the system is defined as a binary variable that takes the value 1 when a pipe fails in the period of study, and 0 otherwise. Secondly, the initial focus of this thesis was extended to multi label classification, which allows predicting more than one output variable at the same time. The aim of, En esta Tesis se aborda el problema de la aparición de roturas o fallos inesperados en las tuberías que componen las redes de distribución de agua. Concretamente, se busca predecir dichas roturas utilizando técnicas basadas en el aprendizaje automático, del inglés machine learning. Tras un análisis exhaustivo de la literatura existente sobre el tema, se detecta que, aunque ya existen estudios que proponen ciertas técnicas de machine learning para el propósito anteriormente descrito, es una temática reciente que aún no ha sido desarrollada en su totalidad. Por ello, este trabajo propone distintos modelos de machine learning, algunos de los cuales no han sido aplicados a la problemática de estudio hasta la fecha, y analiza los aspectos más significativos del procesamiento de los datos y de la evaluación de los resultados. En el desarrollo de un sistema de machine learning tiene especial importancia la forma y características de los datos a utilizar. En este trabajo, se dispone de una extensa base de datos de la red de abastecimiento de agua de Sevilla, la cual ha sido cedida por la Empresa Metropolitana de Abastecimiento y Saneamiento de Aguas de Sevilla (EMASESA), compañía que gestiona dicha red. La base de datos consta de un histórico de roturas de siete años, de 2012 a 2018 inclusive, e incluye diversas variables relacionadas con cada una de las tuberías que forman sus más de 3800 kilómetros de red. Como primer enfoque, se decide explorar la predicción de fallos en las tuberías con un horizonte temporal de un año, dado que las compañías generalmente planifican las tareas de mantenimiento y reposición de la red a un año vista. Por ello, y de acuerdo a las características del problema y a los datos disponibles, se proponen las siguientes técnicas de machine learning: el análisis discriminante, la regresión logística, las máquinas de vector soporte, los bosques aleatorios, las redes neuronales y la lógica difusa evolutiva. Todas estas técnicas tienen la capacidad de a

This PhD thesis addresses the problem of the appearance of unexpected pipe failures in water distribution networks. Specifically, it seeks to predict such failures using machine learning based techniques. An in depth literature review on the subject informs that although there are studies that have tested certain machine learning techniques for the aforementioned purpose, this is a novel issue that has not been fully explored yet. Consequently, this work proposes several machine learning models, some of which have not been applied to this problem before and analyses the most significant aspects of data processing and evaluation of the results. The nature and characteristics of the data are key points on the design of a machine learning system. For the development of this thesis, the company that manages the water distribution network of Seville (Spain) called EMASESA has provided an extensive database. Concretely, the database consists of a seven year pipe failure history, from 2012 to 2018, and includes various factors related to each of the pipes that compose the more than 3800 kilometres network. The first strategy has been to forecast pipe failures one year in advance, since companies generally decide their maintenance and replacement plans annually. Therefore, and according to the characteristics of the problem and the available data, the following machine learning techniques are proposed: discriminant analysis, logistic regression, support vector machines, random forests, artificial neural networks and evolutionary fuzzy logic. All these models can work as classifiers, being the main part of a supervised classification machine learning system. In this case, the output of the system is defined as a binary variable that takes the value 1 when a pipe fails in the period of study, and 0 otherwise. Secondly, the initial focus of this thesis was extended to multi label classification, which allows predicting more than one output variable at the same time. The aim of, En esta Tesis se aborda el problema de la aparición de roturas o fallos inesperados en las tuberías que componen las redes de distribución de agua. Concretamente, se busca predecir dichas roturas utilizando técnicas basadas en el aprendizaje automático, del inglés machine learning. Tras un análisis exhaustivo de la literatura existente sobre el tema, se detecta que, aunque ya existen estudios que proponen ciertas técnicas de machine learning para el propósito anteriormente descrito, es una temática reciente que aún no ha sido desarrollada en su totalidad. Por ello, este trabajo propone distintos modelos de machine learning, algunos de los cuales no han sido aplicados a la problemática de estudio hasta la fecha, y analiza los aspectos más significativos del procesamiento de los datos y de la evaluación de los resultados. En el desarrollo de un sistema de machine learning tiene especial importancia la forma y características de los datos a utilizar. En este trabajo, se dispone de una extensa base de datos de la red de abastecimiento de agua de Sevilla, la cual ha sido cedida por la Empresa Metropolitana de Abastecimiento y Saneamiento de Aguas de Sevilla (EMASESA), compañía que gestiona dicha red. La base de datos consta de un histórico de roturas de siete años, de 2012 a 2018 inclusive, e incluye diversas variables relacionadas con cada una de las tuberías que forman sus más de 3800 kilómetros de red. Como primer enfoque, se decide explorar la predicción de fallos en las tuberías con un horizonte temporal de un año, dado que las compañías generalmente planifican las tareas de mantenimiento y reposición de la red a un año vista. Por ello, y de acuerdo a las características del problema y a los datos disponibles, se proponen las siguientes técnicas de machine learning: el análisis discriminante, la regresión logística, las máquinas de vector soporte, los bosques aleatorios, las redes neuronales y la lógica difusa evolutiva. Todas estas técnicas tienen la capacidad de a

The water supply networks of many countries are experiencing a drastic increase in the number of pipe failures. To reverse this tendency, it is essential to optimise the replacement plans of pipes. For this reason, companies demand pioneering techniques to predict which pipes are more prone to fail. In this study, an Artificial Neural Network (ANN) is designed to classify pipes according to their predisposition to fail based on physical and operational input variables. In addition, the usefulness and effectiveness of two sampling methods, under-sampling and over-sampling, are analysed. The implementation of the model is done using the open-source software Weka, which is specialised in machine-learning algorithms. The system is tested with a database from a real water network in Spain, obtaining high-accurate results. It is verified that the balance of the training set is imperative to increase the predictions’ accurateness. Furthermore, under-sampling prioritises true positive rates, whereas over-sampling makes the system learn to predict failures and non-failures with the same precision.

El presente trabajo consiste en un estudio del reparto urbano de mercancías en la ciudad de Sevilla. El objetivo de este trabajo es crear una base de datos completa y fiable de las características que rodean a los repartos estudiados. Tras ello, se comentarán los problemas observados durante el trabajo de campo y se analizarán los datos obtenidos en busca de posibles relaciones entre características de los repartos y tipos de estacionamientos usados en los mismos. La recogida de datos se ha basado principalmente en la observación directa, pero también en entrevistas con los trabajadores de los negocios o transportistas. Con esto, se busca que los datos obtenidos sean fiables y reflejen la realidad de cómo se realizan los repartos urbanos de mercancías., This work is focused on the study of the urban distribution of goods in the city of Seville. The objective of this work is to create a complete and reliable database of different characteristics concerning the studied deliveries. After that, the detected problems during the field work will be discussed and the obtained data will be analyzed in order to find possible relationships between the characteristics of the deliveries and the kind of parking used. Data collection is based on direct observation mainly, but also on interviews with business workers or carriers. Thus, we aimed to obtained first-hand data, reflecting the urban goods deliveries in a reliable way.

In this study, an evolutionary fuzzy system is proposed to predict unexpected pipe failures in water supply networks. The system seeks to underpin the decisions of management companies regarding the maintenance and replacement plans of pipes. On the one hand, fuzzy logic provides high degrees of interpretability over other black box models, which is requested in engineering application where decisions have social consequences. On the other hand, the genetic algorithm helps to optimize the parameters that govern the model, specifically, for two purposes: (i) the selection of variables; and (ii) the optimization of membership functions. Data from a real water supply network are used to evaluate the accuracy of the developed system. Several graphs that depict the ranking of pipes according to their risk of failure against the network length to be replaced support the choice of the most successful model. In fact, results demonstrate that the annual replacement of 6.75% of the network length makes it possible to prevent 41.14% of unexpected pipe failures

El presente Trabajo Fin de Máster se centra en la elaboración de un modelo de localización de instalaciones semi-deseables, con el que se pretende realizar el análisis de la ubicación de instalaciones llamadas Ecopuntos en los distintos barrios de la ciudad de Sevilla, y compararlo posteriormente con la ubicación ya existente de estas instalaciones en la ciudad. Para ello, se ha desarrollado un modelo de Programación Lineal Entera Mixta (MILP) con un enfoque bi-objetivo, cuya función objetivo está compuesta por un objetivo relacionado con el coste de inversión de las instalaciones y otro que hace referencia a la minimización del nivel de insatisfacción que genera la instalación en los vecinos asignados a ésta, comprendido este último por factores tanto de aceptación como de rechazo. Para llevar a cabo la resolución del modelo, éste se ha implementado en el optimizador Gurobi, habiendo realizado el código en Python, del cual se han extraído los resultados para su posterior análisis. La resolución del modelo proporciona el número de Ecopuntos que deben abrirse, la ubicación de cada uno de ellos en los barrios de la ciudad de Sevilla, los barrios asignados a cada instalación y los valores de coste e insatisfacción de cada solución obtenida. Con los resultados obtenidos para la ubicación de estas instalaciones en los barrios de la ciudad de Sevilla, se demuestra que es posible aplicar el enfoque bi-objetivo en la resolución de problemas reales., This thesis is focused on the development of a decision model for locating semi-desirable facilities, with the aim of analyzing the location of Ecopuntos facilities in different neighborhoods of the city of Seville and then comparing it with the existing location of these type of facilities in the city. In order to do that, a Mixed Integer Linear Programming (MILP) model has been developed with a bi-objective approach, whose objective function is composed of an objective related to the investment cost of the facilities and another that refers to the minimization of the dissatisfaction level caused by facilities amongst residents assigned to them, consisting of both acceptance and rejection factors. In order to carry out the model resolution, it has been implemented in the mathematical programming solver Gurobi Optimizer, the code has been created in Python, from which results have been extracted for their future analysis. The resolution of the model provides the number of Ecopuntos to be opened, the location of each one of them in different neighborhoods of the city of Seville, neighborhoods assigned to each installation and the values of cost and dissatisfaction level for each solution obtained. The results obtained for the location of these installations in the neighborhoods of the city of Seville show that it is possible to apply the bi-objective approach to a real planning scenario.

Intermodal freight transport consists of using different modes of transport without changing the load unit. This results in a significant reduction in the time that goods spend at intermodal terminals, where transshipment takes place. Drayage refers to the transport of freight on trucks among intermodal terminals, depots, customers and suppliers. In spite of the fact that drayage only represents between 5 and 10 percent of total distance, it may amount up to more than 30 percent of the total costs. The aim of this work is to study drayage operations. First, an extensive literature review is undertaken. Since the intermodal transport chain can become more efficient by means of a proper organisation of the drayage movements, the optimization of the daily drayage problem has been identified as one of the main ways of reducing the drayage cost and improving intermodal operations. On this problem, the lack of a common benchmark has hindered reaching further conclusions from all the research carried out. Therefore, this paper proposes a common framework and presents a generalized formulation of the problem, which allows modeling most drayage policies, with the limitation of only considering one-container problems. Results show that flexible tasks in the repositioning of empty containers as well as soft time windows can reduce the operating costs and facilitate the management of drayage companies. This work may help consider adequate policies regarding drayage operations in intermodal terminals.

The assessment of indoor thermal comfort in schools has become an essential object of study; however, applying existing thermal comfort criteria would assume children and adults have a similar range of thermal comfort, without considering discrepancies regarding their level of activity or their behavioural adaptation. Therefore, the objective of the present study was to investigate the thermal comfort in a school building based on an adaptive thermal comfort field study in Seville, in the southwest of Spain, during a summer season. In this study, 2 free-running and 1 air-conditioned classroom were analysed; 67 students aged 10–11 years participated and 2010 thermal questionnaires were collected. A discrepancy was observed between the predicted mean vote and the thermal sensation vote, showing the former is not a good predictor of thermal perception. Thermoneutrality was not always the desired sensation for children; a preference for coolness was detected. A neutral temperature was observed at an average indoor temperature of 24–27 °C and a widening in the thermal comfort range was detected compared with international standards. Regarding adaptive strategies, they showed a preference towards opening windows and doors over using fans or changing clothes. The results suggest that the application of the current models for adults would not be suitable for estimating the thermal comfort of children, and these data could be used to promote natural strategies for assessing thermal comfort over conditioning systems in schools, with the aim of both space ventilation and energy efficiency.

The water supply networks of many countries are experiencing a drastic increase in the number of pipe failures. To reverse this tendency, it is essential to optimise the replacement plans of pipes. For this reason, companies demand pioneering techniques to predict which pipes are more prone to fail. In this study, an Artificial Neural Network (ANN) is designed to classify pipes according to their predisposition to fail based on physical and operational input variables. In addition, the usefulness and effectiveness of two sampling methods, under-sampling and over-sampling, are analysed. The implementation of the model is done using the open-source software Weka, which is specialised in machine-learning algorithms. The system is tested with a database from a real water network in Spain, obtaining high-accurate results. It is verified that the balance of the training set is imperative to increase the predictions’ accurateness. Furthermore, under-sampling prioritises true positive rates, whereas over-sampling makes the system learn to predict failures and non-failures with the same precision.

Asset management in hydraulic infrastructures aims for the long-term sustainability of water distribution and wastewater networks. Strategic maintenance planning has thus been deeply analyzed in the literature for indi vidual water and sewer pipes. However, water utilities do not plan and perform replacement activities on in dividual elements, but rather on coherent aggregations of neighboring pipes. We have developed a decision support system (DSS) to help water utilities design intervention programs for hydraulic infrastructures. It in tegrates a two-stage algorithm that groups water supply and sewer pipes into practical and efficient replacement works, based upon their proximity and their priority of renewal. A multi-objective genetic algorithm optimizes the work programs configurations while integrating the water company’s strategic policy into an innovative multi-objective function. We have applied our methodology to a large water company in Spain and illustrated this application with a sensitivity analysis to determine how the company’s strategic criteria influences the resulting work configurations.

In this study, an evolutionary fuzzy system is proposed to predict unexpected pipe failures in water supply networks. The system seeks to underpin the decisions of management companies regarding the maintenance and replacement plans of pipes. On the one hand, fuzzy logic provides high degrees of interpretability over other black box models, which is requested in engineering application where decisions have social consequences. On the other hand, the genetic algorithm helps to optimize the parameters that govern the model, specifically, for two purposes: (i) the selection of variables; and (ii) the optimization of membership functions. Data from a real water supply network are used to evaluate the accuracy of the developed system. Several graphs that depict the ranking of pipes according to their risk of failure against the network length to be replaced support the choice of the most successful model. In fact, results demonstrate that the annual replacement of 6.75% of the network length makes it possible to prevent 41.14% of unexpected pipe failures

Intermodal freight transport consists of using different modes of transport without changing the load unit. This results in a significant reduction in the time that goods spend at intermodal terminals, where transshipment takes place. Drayage refers to the transport of freight on trucks among intermodal terminals, depots, customers and suppliers. In spite of the fact that drayage only represents between 5 and 10 percent of total distance, it may amount up to more than 30 percent of the total costs. The aim of this work is to study drayage operations. First, an extensive literature review is undertaken. Since the intermodal transport chain can become more efficient by means of a proper organisation of the drayage movements, the optimization of the daily drayage problem has been identified as one of the main ways of reducing the drayage cost and improving intermodal operations. On this problem, the lack of a common benchmark has hindered reaching further conclusions from all the research carried out. Therefore, this paper proposes a common framework and presents a generalized formulation of the problem, which allows modeling most drayage policies, with the limitation of only considering one-container problems. Results show that flexible tasks in the repositioning of empty containers as well as soft time windows can reduce the operating costs and facilitate the management of drayage companies. This work may help consider adequate policies regarding drayage operations in intermodal terminals.

The assessment of indoor thermal comfort in schools has become an essential object of study; however, applying existing thermal comfort criteria would assume children and adults have a similar range of thermal comfort, without considering discrepancies regarding their level of activity or their behavioural adaptation. Therefore, the objective of the present study was to investigate the thermal comfort in a school building based on an adaptive thermal comfort field study in Seville, in the southwest of Spain, during a summer season. In this study, 2 free-running and 1 air-conditioned classroom were analysed; 67 students aged 10–11 years participated and 2010 thermal questionnaires were collected. A discrepancy was observed between the predicted mean vote and the thermal sensation vote, showing the former is not a good predictor of thermal perception. Thermoneutrality was not always the desired sensation for children; a preference for coolness was detected. A neutral temperature was observed at an average indoor temperature of 24–27 °C and a widening in the thermal comfort range was detected compared with international standards. Regarding adaptive strategies, they showed a preference towards opening windows and doors over using fans or changing clothes. The results suggest that the application of the current models for adults would not be suitable for estimating the thermal comfort of children, and these data could be used to promote natural strategies for assessing thermal comfort over conditioning systems in schools, with the aim of both space ventilation and energy efficiency.