Translator Disclaimer
28 August 2019 Robust local Spatial Autocorrelation Analysis of Massive Vessel Movements
Myeong-Hun Jeong, Dong Ha Lee, Tae Young Lee, Jung Hwan Lee
Author Affiliations +

Jeong, M.-H.; Lee, D.H.; Lee, T.Y., and Lee, J.H., 2019. Robust local spatial autocorrelation analysis of massive vessel movements. In: Lee, J.L.; Yoon, J.-S.; Cho, W.C.; Muin, M., and Lee, J. (eds.), The 3rd International Water Safety Symposium. Journal of Coastal Research, Special Issue No. 91, pp. 306-310. Coconut Creek (Florida), ISSN 0749-0208.

Automatic identification system (AIS) data provide rich information on maritime spatial planning and monitoring. This study investigates the local spatial clusters of vessel movements based on AIS data. However, the traditional local spatial autocorrelation method, such as the Getis-Ord (fi01_306.gif) statistic, is based on the assumptions that the underlying data are randomly sampled from a normal distribution. Unfortunately, empirical datasets, such as AIS data, frequently violate the normal distribution assumption. This study, therefore, uses robust statistic methods, which are insensitive to skewness and heavy-tailed distribution, to enhance the traditional local fi01_306.gif statistic. The results demonstrated that the robust local fi01_306.gif statistic provides a true reflection of AIS data, compared with the traditional methods under non-normality. The proposed method can be applied to a broad range of maritime monitoring and security, which results in getting a better insight into maritime data.

©Coastal Education and Research Foundation, Inc. 2019
Myeong-Hun Jeong, Dong Ha Lee, Tae Young Lee, and Jung Hwan Lee "Robust local Spatial Autocorrelation Analysis of Massive Vessel Movements," Journal of Coastal Research 91(sp1), 306-310, (28 August 2019).
Received: 9 October 2018; Accepted: 14 December 2018; Published: 28 August 2019

Get copyright permission
Back to Top