Dhanani, A; Vaughan, LS; Ellul, C; Griffiths, S; (2012) From the axial line to the walked line: Evaluating the utility of commercial and user-generated street network datasets in space syntax analysis. In: Greene, M, (ed.) Proceedings: Eighth International Space Syntax Symposium.

PDF Dhanani_etal_2012_sss-8.pdf Download (9MB)

Abstract

Data availability, reliability and cost are some of the most constraining factors in space syntax analysis and wider commercial acceptance. In recent years user-created Volunteered Geographic Information (VGI) that is free to all via the Internet has gained wider acceptance and proven reliability (Haklay, 2010). Furthermore it has the property of being created by the people who inhabit the spaces being mapped; therefore it captures local knowledge and detail to a far greater degree than commercial mapping agencies. From a space syntax perspective it also relates more closely to the pedestrian network as it is used on foot and captures details of pedestrian routes through the urban fabric that other road-centric data sources ignore. This paper demonstrates the methodological approaches and analytic outcomes of a space syntax sensitivity analysis of Open Street Map (OSM) VGI road network data, the UK national mapping agency Ordnance Survey Integrated Transport Network (ITN) road data and a hand-drawn Axial map for four areas within the Greater London Region. The space syntax segment analysis was completed within the Depthmap application. The segment analysis was completed on the ITN model, OSM model and hand-drawn model separately and then it was carried out on a combined model of the ITN and OSM that integrated all the network elements from both. The integration and comparison of the network models was carried out through the usage of a new algorithm currently under development at University College London that identifies and extracts the differences between two line network datasets (Koukoletsos, forthcoming) and standard GIS processing techniques. The space syntax measures were evaluated on four areas in outer London that are the focus of the Adaptable Suburbs project at the Bartlett School of Graduate Studies. The analysis was carried out using length-weighted angular segment and choice analysis at radii 800m, 2000m and n (Turner, 2007). Comparative statistics were then generated for the areas to evaluate the analysis outcomes of the different network models. The London-wide network that was created through the combination of the OSM and ITN networks had a total length of 32,000km representing an increase of approximately 20% over the Ordnance Survey ITN network. The dramatic increase in network length alone demonstrates the divergent realities of the two mapping techniques and the representation of the world that they capture. It is anticipated that the sensitivity analysis will find that there was no significant difference in the global syntax values between the ITN and OSM and Axial models but at the local level the additional network segments for pedestrian routes within the OSM data will provide greater network accuracy and syntax values that model the reality on the ground better than the Ordnance Survey ITN model. Furthermore it captures potential pedestrian routes that are not present in the other data sets. The work carried out seeks to understand whether Volunteered Geographic Information is a viable alternative to official mapping sources when creating models for analysis of small urban areas. If this proves to be the case such data would provide not only a cost effective alternative to commercially produced data but indeed a more reliable network model for the analysis to be carried out. Open source geographic data have the capability to improve and enrich space syntax analysis whilst removing high price barriers that commercial data sources impose.

Download activity - last month

Export as XMLJSONCSV

Download activity - last 12 months

Export as XMLJSONCSV

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item