Data, a new black gold for mobility?



A growing volume of data

Data is everywhere in the field of mobility. Its nature is very diverse depending on the means of collection. The transport offer is understood through the geomatic definition of road networks (OSM, OpenStreetMap), public transport networks (GTFS), the parking offer or the position of EV charging stations (Izivia). Automatic counts provide information on traffic volumes at certain points on the road network, localized surveys provide information on the origins and destinations of individuals. The massive deployment of smartphones and GPS is continually increasing the number of tracking vehicles (FCD, Floating Car Data or FMD, Floating Mobile Data) which allow the geolocation of vehicles, whether personal or belonging to vehicles fleets of new mobility services. This location data itself produces new information such as trafficolor maps (Google Maps) or travel times on routes. Digital models such as trip models can also be generators of data such as origin-destination matrices. These are also fed by socio-economic data like population census data and activities location data. The composition of the vehicle fleet is also valuable information that can be retrieved, for example, using vehicle registration files. The types of vehicles are strongly linked to the level of pollution, which can be measured using atmospheric sensors. That’s a matter of fact, data are as numerous as they are varied and their volume is growing day by day.


Commercial or open data?

Data accessibility is a major issue for mobility players. There are almost as many providers as there are types of data. However data can be segmented into two main categories:


(1) Commercial data: it is produced (collected, archived, processed) and disseminated by private providers. They can be transport infrastructure managers (roads, highways, airports, etc.), public and private operators of transport services (public or on-demand transport, fleets of taxis or self-service bicycles), automobile manufacturers (connected vehicles), telcos (Orange, etc.), digital service operators (Apple, Google, Waze, Coyotte, etc.) or multi-source data aggregators (Tomtom, Here Technologies, etc.). The data is via a paid business model, and from a technical point of view via APIs with a very high quality of service. This can be both real-time access or access to historical datasets.


(2) Open data: that is to say data collected or whose property is often that of public authorities, cities and metropolises, mobility authorities (AOM), State services, and public institution. In France, since the Le Maire law on Open Data, public data is open, which means accessibility for all, and it is free. Since then, the Mobility Orientation Law (article 25) has supplemented this vision by encouraging the opening of mobility data. A new license, the Mobility license, has recently been introduced to regulate the usage of public data in digital mobility services so that they are always more respectful of the general interest.


However, ensuring that all public mobility data is defacto open remains a complex operation. It is important to distinguish between recurring data (whose collection methods are permanent), and occasional data (collected temporarily during major projects or specific events). It is easier for the public players to publish the data that they are handling on a daily basis and which is often integrated into an automated digital process (even if the question of the interoperability of formats arises), while the question of the heterogeneity of formats, the lack of standardization in reference systems and the geographical and temporal diversity of the series of point data raises the question of the economic and collective utility of opening up these data.


Various applications

In all business areas, the enthusiasm for data is real. The mobility sector is no exception to this underlying trend: data is becoming black gold for many uses and applications, and not only in the field of the general public.


(1) Data enables to understand the past, for example by replaying situations for which the information has been recorded. Technological development brings new means to make this data speak, among others of artificial intelligence and data science (eg statistics, machine learning). Classifying (unsupervised machine learning - clustering) enable to extract recurring mobility patterns. Metrics (statistics, KPI) are used to assess past performance to assess the potential for improvement. Statistical analyzes provide information on correlations between datasets, and provide elements to deduce causalities. But, correlation does not imply a causal link!. This is the role of mobility observatories, which are regularly updated by local authorities and their service providers.


(2) Data enables to monitor the current situation. It gives a real time knowledge (snapshot) of what is happening on the transport network or in the mobility service in order to supervise operations.


(3) Data enables to plan the future: for short-term forecasting, generally within a timeframe that can range from 15 min to 3 hours, we use machine learning techniques (supervised machine learning such as neural networks, models regression, categorization), or physical models (flow calculations, simulation). Long-term forecasting empower planning operations such as infrastructure development projects, traffic management strategies, integration of new mobility systems. To study their systemic impacts, mobility professionals often implement mobility models (digital twins).


Intensive but differentiated usage

Today, all players in the mobility sector are data intensive consumers. But the value of data is not the same for all the players depending on their place in the value chain. The following examples are a quick illustration of that diversity:

  • Cities and local authorities, which support local policies driven by their elected representatives.

  • Network operators (road, public transport, etc.), whose monitoring challenges are a daily reality, and whose needs for projects assessment are recurring.

  • Engineering and consulting companies, which assess projects on behalf of project owners and network operators.

  • Business software publishers, as well as researchers, who have to develop efficient and generic models and methods, to improve the processes of all the players in the system.


What challenges to come?

The accessibility and sharing of data does not become a reality through simple statements. Upstream and downstream work is necessary. In terms of access to data, open data portals are multiplying, but the depth of their catalog makes them difficult to understand and complicates the use of the available datasets. Commercial data is expensive and business models are not always easy to understand. Commercial data is already accessible through robust technologies. It now requires support from institutional public players to increase visibility in the ecosystem and also some clarification of pricing policies in order to spread their use.


Open data has a whole regulatory and legal framework for its use. It would now be the time to go a step further with a technical framework to support public data providers in the digitalization of their process to open their data. This apply to the development and implementation of standards adapted to each type of data (as it is already done for example with DATEX 2 for road traffic data, or with GTFS, initiated by Google, for public transport and parking). The construction of an appropriate governance (local, territorial, national) is also a crucial question to capitalize, mutualize and share data between neighboring and complementary stakeholders. This question must also integrate the identification of key public actors to aggregate the databases of local authorities in order to disseminate at a more global scale.


All mobility players have an interest in working efficiently and hand in hand to get optimal value from mobility data. The main challenge for all players remains to be aligned with the collective interest of transitioning towards a more virtuous, more environmentally friendly and more carbon-free mobility.


Wishing to discover how Neovya Hubsim empowers mobility professionals with data analytics ?
Engage conversation with an expert ➜Book a meeting!