A carregar...
Publicação
Análise de padrões dos pedidos de intervenção do Regimento de Sapadores Bombeiros
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 17.58 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
O Regimento de Sapadores Bombeiros (RSB) é um corpo municipal de bombeiros que,
ao contrário dos bombeiros voluntários, são profissionais. Este corpo é responsável por
trabalhos relacionados com o socorro e apoio à população e acções de protecção civil.
O factor tempo é crucial em situações de emergência. Para melhorar a prestação de
serviços de emergência aos cidadãos e para optimizar a distribuição de meios é necessário
descobrir padrões nos dados das ocorrências deste regimento. Os tipos de padrões abor dados podem ser padrões temporais, espaciais e espácio-temporais. Quando os padrões
estiverem definidos é necessário encontrar variáveis que possam explicar esses padrões.
Por norma, estas variáveis são identificadas através de conjuntos de dados externos que
estão correlacionados com os dados originais como potenciais factores explicativos, por
exemplo, dados de meteorologia. Com a descoberta de padrões e variáveis explicativas é
possível estimar cenários de intervenção. Podendo estimar-se estes cenários tornará possí vel melhorar a organização e distribuição dos meios de emergência e, consequentemente,
será possível diminuir o tempo de resposta das entidades competentes.
Com dados disponibilizados pela Câmara Municipal de Lisboa, relativos aos pedidos
de intervenção recebidos por este regimento entre os anos de 2011 e 2018, foram criadas
ferramentas interativas de modo a permitir a deteção de padrões temporais, espaciais
e espácio-temporais e, posteriormente, proceder à explicação dos mesmos. O objectivo
final é disponibilizar ferramentas visuais que contribuam para a deteção e compreensão
dos padrões nas ocorrências para melhorar a gestão operacional dos bombeiros. Foram
utilizadas técnicas de clustering e técnicas de visualização e interação em dados espácio temporais para o desenvolvimento das ferramentas visuais interativas.
Por fim, as ferramentas desenvolvidas foram avaliadas com um grupo de colegas e
professores da faculdade. A avaliação experimental foi realizada remotamente e consistiu
num questionário com um conjunto de questões onde as métricas recolhidas para análise
foram o tempo de resposta e a taxa de sucesso para cada pergunta.
The Sapper Firefighters Regiment is a municipal fire department which, unlike volunteer firefighters, are professional. This department is responsible for relief and support work to the population and civil protection actions. The time factor in an emergency is crucial. To improve the provision of emergency services to citizens and to optimize the distribution of means, it is necessary to discover patterns in the occurrence data of this regiment. The types of patterns covered can be temporal, spatial and spatial-temporal. When patterns are defined, it is necessary to find variables that can explain these patterns. Typically, these variables are identified through external data sets that are correlated with the original data as potential explanatory fac tors, for example, weather data. With the discovery of patterns and the explanatory variables, it is possible to estimate intervention scenarios. Being able to estimate these scenarios will make it possible to improve the organization and distribution of emer gency means and, consequently, it will be possible to decrease the response time of the competent entities. With data provided by the Lisbon City Council, concerning the intervention requests received by this regiment between the years 2011 and 2018, interactive tools were cre ated to allow the detection of temporal, spatial and spatio-temporal patterns and, subse quently, to explain them. The ultimate goal is to provide visual tools that contribute to the detection and understanding of patterns in events to improve the operational manage ment of firefighters. Clustering techniques and visualization and interaction techniques were used in spatio-temporal data for the development of interactive visual tools. Finally, the tools developed were evaluated with a group of colleagues and professors from the college. The experimental evaluation was carried out remotely and consisted of a questionnaire with a set of questions where the metrics collected for analysis were the response time and the success rate for each question.
The Sapper Firefighters Regiment is a municipal fire department which, unlike volunteer firefighters, are professional. This department is responsible for relief and support work to the population and civil protection actions. The time factor in an emergency is crucial. To improve the provision of emergency services to citizens and to optimize the distribution of means, it is necessary to discover patterns in the occurrence data of this regiment. The types of patterns covered can be temporal, spatial and spatial-temporal. When patterns are defined, it is necessary to find variables that can explain these patterns. Typically, these variables are identified through external data sets that are correlated with the original data as potential explanatory fac tors, for example, weather data. With the discovery of patterns and the explanatory variables, it is possible to estimate intervention scenarios. Being able to estimate these scenarios will make it possible to improve the organization and distribution of emer gency means and, consequently, it will be possible to decrease the response time of the competent entities. With data provided by the Lisbon City Council, concerning the intervention requests received by this regiment between the years 2011 and 2018, interactive tools were cre ated to allow the detection of temporal, spatial and spatio-temporal patterns and, subse quently, to explain them. The ultimate goal is to provide visual tools that contribute to the detection and understanding of patterns in events to improve the operational manage ment of firefighters. Clustering techniques and visualization and interaction techniques were used in spatio-temporal data for the development of interactive visual tools. Finally, the tools developed were evaluated with a group of colleagues and professors from the college. The experimental evaluation was carried out remotely and consisted of a questionnaire with a set of questions where the metrics collected for analysis were the response time and the success rate for each question.
Descrição
Palavras-chave
Regimento de Sapadores Bombeiros Análise de Dados Padrões em Dados Dados Espácio-temporais Técnicas de Clustering Visualização Interativa de Dados