Last modified: 2018-05-18
Abstract
Organizing the emergency medical system in a big city is an extremelydifficult task given the huge number of people that everyday pass through the cityarea. In this paper we employ a spatio-temporal process to model the emergencyevent occurrences in Milan. The proposed approach has been found effective inpredicting events through the city area and computationally efficient despite the bigamount of data to be processed.
L’organizzazione di un servizio di emergenza sul territorio risulta essereun compito complesso nelle aree metropolitane come Milano dato l’enorme numerodi persone che vi transitano quotidianamente. In questo lavoro si adotta un modellospazio temporale per rappresentare la dinamica delle chiamate di emergenzasul territorio del capologuo Lombardo. Il metodo adottato si ´e dimostrato essereefficace nel prevedere gli eventi sul territorio comunale nel periodo di tempo consideratoe computazionalmente efficiente nonostante la consistente mole di dati daelaborare.
References
1. Diggle, P.J.: Statistical Analysis of Spatial and Spatio-Temporal Point Patterns, Chapman andHall/CRC, Boca Raton, Florida (2013)
2. Matteson, D.S., McLean, M.W.,Woodard, D.B. and Henderson, S.G.: Forecasting emergencymedical service call arrival rates. Ann. Applied Statistics 5, 1379-1406, (2011)
3. Shen, H. and Huang, J.: Forecasting Time Series of Inhomogeneous Poisson Processes withApplication to Call Center Workforce Managament. Ann. Applied Statistics 2, 601-623,(2008)
4. Wood, S.N.: Generalized Additive Models: An Introduction with R, Chapman and Hall/CRC,Boca Raton, Florida (2017)
5. Zhou, Z. and Matteson, D.S.: Predicting ambulance demand: A spatio-temporal kernel approach.ACM SIGKDD. ACM 2297-2303, (2015).
6. R Core Team: R: A Language and Environment for Statistical Computing. Vienna, Austria,(2017).