Geospatial mining of massive AIS data for strategic management of maritime navigation safety: Methodological and policy considerations

Rawson, Andrew, David, Brito, Mario and Sabeur, Zoheir (2021) Geospatial mining of massive AIS data for strategic management of maritime navigation safety: Methodological and policy considerations. Royal Institute of Navigation Conference 2021, Virtual. 15 - 18 Nov 2021. (In Press)

Record type: Conference or Workshop Item (Paper)

Abstract

A spatial model of maritime risk would be useful to navigational authorities in accident prevention and waterway management. Mapping maritime incident rates, the frequency of accidents per vessel movement within an area could provide such an evidence base. However, aggregating spatial data is subject to challenges associated with the Modifiable Areal Unit Problem that can influence the validity of any outputs and has not been well explored within the context of navigation safety. This paper provides a framework for calculating incident rates from massive vessel traffic and accident databases and contrasts the impact of different spatial data structures and resolutions on the results. The results show that the distribution, correlations and the underlying incident rates varies significantly depending on which spatial data structure is utilised. This paper provides a discussion of these challenges, highlighting their possible impacts on maritime risk analysis, and providing some possible solutions to address them.

Text

AR_RIN2021_v2 - Accepted Manuscript

Restricted to Repository staff only

Request a copy