Abordagem para otimização do processo de alocação de viaturas policiais utilizando registro de ocorrências criminais para suporte no processo decisório gerencial no âmbito da segurança pública

• Ano de defesa: 2022
• Autor(a) principal: Rocha, Gláucio Bezerra
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Informática; Programa de Pós-Graduação em Informática; UFPB
• Palavras-chave em Português: Informática; Clusterização - Inteligência policial; Segurança Pública - Ocorrências policiais; Viaturas policiais - Problema de localização; Computing; Clustering - Police intelligence; Public security - Police occurrences; Police vehicles - Location problem; Análise criminal; Series temporias; Criminal analysis; Temporary series; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/26689

Considering public security in Brazil as an area of great importance in today's society, being a critical sector in the Brazilian states and directly linked to the greatest legal asset of human beings, life, in addition to heritage, it is necessary to carry out a study that allows mitigating the increase in crime through the use of police intelligence. In an analysis carried out within the scope of public security in the State of Paraíba, one of the problems identified and brought to this work was the poor distribution of police vehicles throughout the geographic area of the cities, causing delays and inefficiencies when serving the population. Currently, police vehicles are distributed in fixed bases located in neighborhoods, where practically each neighborhood has some vehicles available to serve that area, and it is clear that the current allocation of these patrols does not consider some characteristics of the criminal phenomenon to define their locations, the example of number of existing occurrences and distance from places with higher crime incidence. Considering the entire scenario of the problem presented, this work proposes a process of clustering areas through records of police occurrences, to later predict the amount of crimes in these identified regions, and finally, prescribes a solution through the resolution of the Location Problem of Maximum Qualified Coverage adapted to criminal data.