Potentials of Bicycle Infrastructure Data Lakes to Support Cycling Quality Assessment

Abstract

A data-driven quality assessment of bicycle infrastructure is necessary in times of crisis to support the decision-making process in cycling promotion. The INFRASense project was initiated to support the scoring of bike paths by providing new crowdsourcing data that is combined with other relevant data sources (traffic amount, accidents, citizen reportings etc.). The storage and processing of heterogeneous bike infrastructure data may be a challenge. With its flexibility a Data Lake could be an alternative to the traditional Data Warehouse. In the first step the paper gives an overview about data-driven initiatives in the use-case of bike infrastructure quality assessment and the recently started research project INFRASense. We will provide an overview about data sources that may potentially be included into the data driven quality assessment. Big Bicycle Data is available in many different structures and formats (CSV, XML, SHP etc.). In the second step the concepts of Data Lake and Data Warehouse are introduced. The benefits and weaknesses of these two solutions are shown followed by a discussion about which one of these is the best concept for storage, processing, and analysis of heterogeneous bicycle infrastructure data. In the last step we are providing an outlook how an efficient bicycle infrastructure data management system could be implemented.