Spatial decision model for urban planning

• Ano de defesa: 2023
• Autor(a) principal: ROSA, Amanda Gadelha Ferreira
• Orientador(a): MOTA, Caroline Maria de Miranda
• Tipo de documento: Tese
• Tipo de acesso: Acesso embargado
• Idioma: eng
• Instituição de defesa: Universidade Federal de Pernambuco
• Programa de Pós-Graduação: Programa de Pos Graduacao em Engenharia de Producao
• País: Brasil
• Palavras-chave em Português: Engenharia de Produção; Decision model; Urban planning; Spatial decision-making
• Link de acesso: https://repositorio.ufpe.br/handle/123456789/49943

The decision-making process is an innate task for human beings, and since all choices and actions are based on preferences, decisions are naturally made. However, there are more complex decisions that require the use of Multiple Criteria Decision Making/ Analysis (MCDM/A). This thesis presents a collection of articles based on the use of statistical, optimization, and multi- criteria methods for urban planning regarding spatial decision-making. Particularly, we propose the assessment of attractiveness, connectivity, vulnerability to crime and exploration of the role of attractiveness and connectivity in crime event. For this, we used multiple data sources (Brazilian Institute of Geography and Statistics (IBGE), Brazilian National Civil Aviation Agency (ANAC), Brazilian National Telecommunications Agency (ANATEL), Brazilian Central Banking (BCB), OpenStreetMaps (OSM), Google Maps and crime data) which were cleaned and preprocessed to select criteria to achieve these objectives. Utilités Additives Discriminantes (UTADIS) and Dominance-based Rough Set Approach (DRSA) are MCDM/A methods. Through UTADIS, we found that almost 86% of municipalities in Pernambuco are classified as very low attractive, which can alert policymakers to meet population demands. In order to reveal the vulnerability of an area in a city in the state of Pernambuco, Brazil, we used DRSA and found that the presence of at least 15 restaurants can lead to a Census tract (CT) being classified as very highly vulnerable. The results also demonstrated pessimism in relation to vulnerability by indicating the evaluation of areas as more vulnerable than they really are. Regarding the connectivity, we proposed the elucidation of logistics terminals in individual perception, once the connectivity can be measured through the data of connectivity, the information concerning the coverage area and the flows between logistics terminal were considered as factor of contribution in preference analysis, Goal Programming (GP) and Linear Programming (LP) were considered for this objective. Lastly, the exploration of crime events based on attractiveness and connectivity outputs analysis revealed that even during the COVID-19 pandemic, the concentration of robberies remained in the same area, and both attractiveness and connectivity are significant in crime patterns. Thus, this thesis presents different approaches to support urban planning and regional development.