Using Inland-AIS data for estimating passenger transport on the Danube
Objectives and implementation
While Statistics Austria has collected data on freight transport on the Danube on a monthly basis since 1965, there have been no further surveys on passenger transport after changes of the legal basis applying from the reference year 2003. Lately, the importance of the Danube for tourism purposes in the context of international river cruises or day-trips has been steadily increasing and the number of passengers transported on the Austrian section of the Danube has become relevant again.
Statistics Austria analysed the feasibility of a survey for passenger transport on the Austrian part of the Danube using new data sources and innovative methods to reduce the burden on respondents as much as possible. This study was carried out as part of a grant project funded by Eurostat.
To gain results for cruise companies, an estimation model was developed using AIS data (Automatic Identification System) and information on vessel itineraries obtained via web scraping. Data from locks and landing stages, information on vessel types and their capacities as well as the results of a national survey by the national waterway operator viadonau were also included.
Innovation within the project
Innovation is achieved by largely using already existing secondary data to estimate passenger transport carried out by cruises on the Austrian section of the Danube:
- AIS transponder data: In inland navigation, the Inland Automatic Identification System (AIS) is used for the automatic identification, tracking and tracing of vessels. The AIS transponder on board of a vessel enables the exchange of information relevant to the positioning and identification of vessels between vessels via radio technology. Each vessel equipped with an Inland AIS transponder sends static (e.g. vessel number, name), dynamic (e.g. position, speed) and voyage-related (e.g. destination, estimated time of arrival) data.
- Web scraping of timetables: With AIS transponder data alone, it is not possible to determine the place of embarkation or disembarkation for cross-border cruises. Therefore, the timetables of cruise companies were linked with the AIS transponder data of the vessels in order to obtain more information on the itineraries of the vessel. Web scraping methods were used to generate these timetables automatically by systematically extracting information from the html-documents of the cruise companies’ websites.
In addition to these two data sources, data from locks and landing stages, information on vessel types and their capacities as well as the results of a national survey by viadonau were used for the estimation model
Interpretation of the results
The results are designated as experimental because the data are not collected from respondents, but are largely based on estimates by linking several secondary statistical sources. Furthermore, the results are only available for the reporting year 2019. Long time-series analysis, e.g. for quality purposes, are not possible as currently comparative data are not available.
Further information, project results