Advanced optimization and data modeling techniques to improve accesibility and thermal comfort in urban plannning
- Autores: Iñigo Delgado Enales
- Directores de la Tesis: Javier del Ser Lorente (dir. tes.), Patricia Cristina Molina Costa (codir. tes.)
- Lectura: En la Universidad del País Vasco - Euskal Herriko Unibertsitatea ( España ) en 2024
- Idioma: inglés
- Programa de doctorado: Programa de Doctorado en Tecnologías de la Información y Comunicaciones en Redes Móviles / Mobile Network Information and Communication Technologies por la Universidad de A Coruña; la Universidad de Cantabria; la Universidad de Oviedo; la Universidad de Zaragoza y la Universidad del País Vasco/Euskal Herriko Unibertsitatea
- Calificación: Sobresaliente Cum Laude
- Materias:
- Enlaces
- Tesis en acceso abierto en: ADDI
- Resumen
In the 21st century, urban areas have established themselves as the most common form of human settlement, far outnumbering rural areas in terms of population. The trend for cities is that their population will continue to grow, until they account for 70\% of the world's population in 2050. The strategic role that cities play and will play for society in the future is so great, that since 2015, this importance is reflected in one of the 17 Sustainable Development Goals (SDGs) of the United Nations, which aims to make cities more inclusive, safe, resilient and sustainable.
The impact of cities and the decisions made when planning them, not only affect the urban area in question but also have a global impact. One example of the wide impact of urban decision planning is its contribution to pollution in urban areas, further exacerbating climate crisis. Achieving a more sustainable city, that is both resilient and accessible to its inhabitants, depends on proper urban planning. Urban planners usually face complex problems, as several factors come into play. This is why decision support systems play an important role in facilitating the planning and design of more sustainable and livable cities. These systems, which are based on digital tools such as Geographical Information System (GIS) or 3D modeling, can help to analyze and visualize data, evaluate different scenarios and strategies, and foresee the possible consequences of urban planning decisions. In the last decades, within the new digital era with more access to loads of data sources, Artificial Intelligence (AI) is postulated as the new leading support tool in urban-decision making processes. This Thesis explores the possibilities of different AI techniques for urban planning from different perspectives.
In the first part of the Thesis, accessibility and urban comfort are addressed. Specifically, the Thesis proposes a new decision-support system based on multi-criteria optimization combined with a geo-referenced graph model. The framework makes it possible to design urban projects that improve accessibility and reduce exposure to noise and/or air pollution through the installation of urban elements (ramps, escalators, ...) taking into account the total economic cost of the installation. In this way, urban planners are provided with a new tool that takes into account the simultaneous consideration of different objectives in an urban planning problem, thus facilitating the decision on where to undertake cost-effective actions in public spaces.
The second part of the Thesis examines modeling problems that focus on the estimation of heat stress in cities due to climate change. In this second research contribution, the usefulness of image segmentation techniques for correlating spatial and meteorological data with street-level air temperature is explored. The ability to estimate temperatures with a high degree of accuracy allows for the identification of the highest priority areas in cities where urban improvements need to be made to reduce thermal discomfort. The air temperature at street level is estimated both spatially and temporally for a specific use case, and compared with existing, well-established numerical models. Based on the obtained results, neural networks are postulated as a faster and less computationally expensive alternative to the aforementioned numerical modeling counterparts.
Overall, the outcomes derived from this Thesis contribute to the development of an embryonic research area --AI-based urban modeling-- that in the future will enjoy relevant importance. The contributions herein reported expose with evidence the suitability of AI techniques for effectively solving urban planning problems that underly the design of safer and friendlier city ecosystems.
