Nearly everybody calls daily with a mobile phone, sends texts or uses the mobile Internet. In an interview, Norbert Weber, data expert at Motionlogic, explains how these activities can be transformed into valuable information for city and traffic planning.
Compass: What particular type of data does Motionlogic offer?
Norbert Weber: Motionlogic’s data stock is based on anonymous swarm data from Deutsche Telekom’s mobile phone network. This means that the statistical evaluations are based on a solid foundation of approximately 40 million end user devices. This way it is possible to map movement streams – for example, origin-destination matrices – much more precisely and in more differentiated fashion than was previously possible with other methods. This data is available to us anonymously and is only evaluated in groups, so that the strict data protection requirements are always considered during the evaluation.
Compass: For which modes of transport are the data relevant?
Norbert Weber: The advantage of the mobile phone data is that trips with all modes of transport are captured – regardless of whether car, train or bicycle. The sole criterion is that the users carry a mobile phone with them. Currently, we disclose the traffic movements in aggregated form, which means that the modes of transport are not distinguished. In the future, we also want to disclose these separately so that still more precise evaluations are possible.
Compass: What possible uses are there for this data?
Norbert Weber: We use the data first and foremost to create origin-destination matrices. What is interesting here is that trips in the evening and at the weekend flow into the evaluation and it is possible to detect differences depending on the weather and season. In addition, however, traffic and passenger counts can be done, for example in the tube or train. All evaluations are also possible looking back at the past.
Compass: Which new areas of application will there be now due to the cooperation with PTV?
Norbert Weber: PTV is a transport modelling specialist. We can only map the actual state, and sometimes there are limitations on this. By integrating traffic movements from the mobile phone network into a transport model, the statements become predictive. Simulations of changes in the network or public transport schedules can thus be played through. Thanks to the combination of a well-founded database with a model approach, it is possible to answer the relevant questions in the traffic sector.
Compass: Which concrete examples are there of how the data is already being used in the traffic and mobility sectors?
Norbert Weber: In the course of an initial pilot project with PTV, we have generated an origin-destination matrix for the city of Karlsruhe. Here, we have evaluated the traffic movements in a period of six weeks between 136 traffic zones. The origin-destination matrix generated this way has been incorporated into a transport model for the city, which is used for the city’s traffic management.
Compass: What hurdles must still be overcome?
Norbert Weber: In recent months, we have invested a lot of energy into a better database. Thus today, we are able to evaluate much denser data than we could just a few months ago. The next step is the enhancement of the algorithms for interpretation of this quantity of data. Our goal is to be able to distinguish each mode of transport if possible and also to be able to count passengers in trains and buses. We will be working on this in the next few months.
Compass: Where do you see the greatest potential? What are your expectations for the future?
Norbert Weber: In the short term, we see the greatest potential in transport planning-related issues; however, we also see a lot of demand for passenger counting in the public transit sector. Technically, it is also possible to evaluate the data nearly in real time. Thus, new applications in the traffic management sector will be possible. In this context, we are interested in the feedback from cities and communities.
Compass: How is data privacy guaranteed?
Norbert Weber: Data privacy is a central concern for us. We have coordinated the type of data anonymization intensively with the German Federal Office for Data Privacy and Freedom of Information (BfDI). The underlying data includes no person-related characteristics such as names or telephone numbers, and it is fragmented. Another essential element is that only groups with a threshold value of 30 are evaluated. The German Federal Office for Data Privacy has confirmed this approach.