All companies on the webAI platform are precisely located at house number level (geocoded). Based on this precise location information, the companies are assigned micro-geographical location factors.
What do we measure?
The Location Agent was developed by ISTARI.AI to quantify information about company locations. The locations of companies have an enormous influence on the infrastructural, social and cultural connectivity of the companies and their employees. Thus, policy makers at the regional, national, and multinational level want to promote economic growth by developing the right location factors to create a beneficial environment for firms.
We use data from OpenStreetMap (OSM) as a high-resolution geodata source to evaluate the location of each individual company in extreme detail. OSM has global coverage and has been used and validated in scientific research for nearly two decades. ISTARI.AI is now using this data set for the first time to evaluate companies in terms of their location. For this purpose, we have developed various scores for the different dimensions of the quality of a site. These are a “Transport Score”, “Recreational Score”, “Cultural Score” and “Leisure Score” described below.
How do we measure?
The valuation of each company is based on its exact location. Through this location, we relate the company to the local infrastructure mapped in the OSM data. For this purpose, we either measure the distance to the nearest relevant object (for example, the nearest airport) or we count the number of relevant objects in an empirically determined radius (for example, all cultural institutions within a 2km radius).
cultural_score: Number of cultural facilities (e.g. cinemas, museums, churches) within 2km. Higher values indicate a higher level of local cultural offerings.
leisure_score: Number of leisure facilities (e.g. restaurant, cafés, clubs) within a radius of 2km. Higher values indicate a higher level of local leisure facilities.
recreational_score: Number of recreational facilities (e.g. parks, playgrounds, sports halls) within a radius of 2km. Higher values indicate a higher level of local recreational facilities.
transport_score: Number of public transport stops weighted by capacity within a radius of 2km. Higher values indicate higher local public transport capacities.
nearest_airport: Distance in meters to the nearest airport. Lower values indicate closer nearest airport.
nearest_motorway: Distance in meters to the nearest motorway access. Lower values indicate closer nearest motorway access.
nearest_university: Distance in meters to the nearest university. Lower values indicate closer nearest university.
How do you interpret the data?
The OSM scores are theoretically not upper-limited. Their significance arises from the comparison of companies. For example, a company located in a rural area may have a (public) transport score of 3.0 but a nearest motorway distance of 3,000 (meters). Another company located in the centre of a large city with many bus lines could have a
transport score of 250.0 and a nearest motorway distance of 7,000.
On average, European companies have a transport score of 129.0, with values ranging between 0 and 2,149. Approximately 5.2% of all the companies in our database have a transport score of 0, i.e. no public transport within 2km of their firm location. Similarly, the mean recreational score is 50.3, the average cultural score 11.2 and the mean for the cultural score at 109.0. 20.9% all firms are located further than 2km from the nearest cultural facility.
In contrast to simpler, binary classifications (e.g. “well connected” / “poorly connected”), these continuous scores at company level offer the possibility of making
microgeographical statements about company locations. Users of the Location Analysis Agent can thus easily determine for themselves which levels of a certain
location factors should apply to a company in order for it to be relevant to them.
How do we ensure the data quality?
OSM is an open-source data source that can be edited by anyone. Nevertheless, numerous scientific studies have shown that the data quality is sufficient for almost all scientific analyses and often even surpasses public or commercial data sets. This is especially the case for urban areas in Central Europe.