Extracting Relevant Points of Interest from Open Street Map to Support E-Mobility Infrastructure Models

Abstract

In addition to commercial geodata, Volunteered Geographic Information (VGI) is gaining more and more importance in research. Platforms like Open Street Map (OSM) meanwhile provide an enormous amount of geodata. At the same time, however, new questions arise regarding the quality and the possibilities of using OSM data for research matters. Therefore, in the field of spatial planning, data require
further validation processes and data cleaning frameworks. This paper presents a method of data processing in the context of electric mobility (e-mobility) research with a focus on a charging station placement model. The presented methodology is divided into pre-validation, to gather the relevant data set, and data processing, that specifies the relevant Points of Interest (POI) for further research by deleting all possible complications arising in OSM data. The validation process is customized to the model that determines the demand of electric charging by categorizing POIs into the four time slots living, work, shopping and recreation. By processing data in the presented way, the electric vehicle charging model is filled with improved input data, which allows to reduce the bias associated to the particularities of the OSM production process. A case study in the Bavarian-Czech border area demonstrates that the error correction rate through the model is at about 10%.