Modelling and Linking Accessibility Data in the Public Bus Network

Many organizations and public administrations are currently working to open their data for the general use, in the context of the Linked Open Data (LOD) initiative. This emphasizes the need to publish available data in a structured format, so that they can be accessed and externally processed in order to maximize their utility for citizens. This also poses a general challenge in the area of information technologies, as it requires being able to integrate data from heterogeneous distributed sources, provided in a compatible format. We consider the specific domain of public transport networks, where the information has obviously a public interest. The processing of these data can be used for a wide variety of applications: route planning systems are a classic example. However, most of the existing approaches in this domain fail to provide specific means to deal with accessibility data, i.e. the information which is relevant for people with special mobility needs, especially when it is obtained from different transit networks. In this context, the LD approach is particularly useful, due to the complexity of those relationships and their inherently graph-oriented nature. In this work we describe the process we use to define this information and to make it available in the context of the CoMobility project. First, we define a conceptual model, supported by the main reference data models in the transport domain: Transmodel and IFOPT, and emphasizing the role of accessibility. We transform it into an ontology, in order to combine data from diverse sources; and then such concrete data are captured using the schema defined by this ontology. In this context, we use actual data from the public bus network in the City of Madrid. Afterwards we are able to define and build new mobility services, e.g. to automatically decide if a certain bus route is accessible or not, considering any potential transfers in this route. Finally, we are able to publish the obtained information, albeit partially, preserving the LOD spirit.