Crowd sourced vs centralized data for transport planning: a case study of bicycle path data in the UK

Abstract

This paper seeks to test the hypothesis that distributed, user-contributed 'crowd sourced' GIS data will eventually supercede the traditional centralised geographic data model. The empirical basis used to explore this question is a couple of national-level datasets on a specific topic: bicycle paths in the UK. Representing the crowd-sourced data model, Open Street Map data was downloaded as a subset of Planet.osm. Representing the centralised model, we used the Ordnance Survey's (OS) Urban Paths data product, which provides a national dataset of dedicated bicycle and walk ways across the UK. To assess the quality of each dataset, an array of tests was used. These ranged from narrow tests of accuracy against aerial photography for contiguous stretches of the route, to more subjective tests of usability and practical utility. Overall it was found that the crowd-sourced data model outperformed the OS approach. However, it must be noted that this is a very niche area with a strong user community. If the crowd-sourced data model is to triumph in more mainstream areas it needs to ensure much greater community 'buy-in', for example through compulsory engagement with Open Street Map for educational and citizenship purposes at school