Open AccessForecasting the Number of Passengers from Bakauheni Port during the Sunda Strait Tsunami Period Using Intervention Analysis Approach and Outlier Detection
Author(s) -
Dani Al Mahkya,
Dian Anggraini
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/537/1/012009
Subject(s) - cruise , port (circuit theory) , period (music) , outlier , time series , geography , geology , computer science , statistics , engineering , oceanography , mathematics , artificial intelligence , physics , acoustics , electrical engineering
The purpose of this study is to model and predict the number of ferry passengers from the Bakauheni port during the Sunda Strait Tsunami period. The Bakauheni-Merak crossing route is one of the sea crossing nodes serving crossings from Sumatra to Java and vice versa. The cruise line with a route length of 27.75 km serves both passenger and vehicle crossings. This route is the main route of logistics distribution from Java to Sumatra and vice versa. The tsunami that happened in the Sunda Strait on 22 December 2018 indirectly impacted the Sunda Strait sea crossing node especially the Bakauheni-Merak route. Methods that can be used to analyze the external impact of an event that causes changes in data patterns are Intervention Analysis and Outlier Detection. Both approaches are a type of Time Series method based on time series data. Therefore, this method can be used to model and predict the number of ferry passengers from the Bakauheni port during the Sunda Strait Tsunami period. If observed visually based on time series data patterns, the number of passengers from the port of Bakauheni has decreased for 2 days after the Tsunami. However, the number of passengers increased again on the Christmas holiday on December 25, 2018 or 3 days after the Tsunami. The tsunami that happened close to the Christmas holidays 2018 gave the results of initial identification, namely the existence of other events that caused changes in the pattern of passenger data. The event was a Christmas holiday on December 25, 2018. This research is expected to be able to capture and analyse phenomena that happen in a series of data during the Sunda Strait Tsunami period.