Abstract

In this paper, a preliminary method is proposed to forecast traffic accidents based on Big data analytics. The study is conducted using the traffic accident records from Bogot´a, Colombia from December 21, 2006 to April 29, 2017. The accidents are visualized considering Spatial-temporal modelling. Then, a characterization and prediction of accidents and the level of damage are done. For the problem, the computation of the Inverse Distance Weighted (IDW) Interpolation with the Radial Basis Function (RBF) in R and the Kriging model are implemented. The preliminary results pretend to support decision making for anticipating possible disruptions of the last-mile supply chain.