To better organize trips, we predict them!

But where are the people going? At what time of the day ? For which motive ? What is the ecological impact of their movements? Answering these questions is the party of the start-up Entropy. Created in 2019, this start-up based in Versailles, in the Yvelines, provides mobility plans to transport operators, design offices, engineering and consulting firms, and sometimes directly to communities, such as the Yvelines departmental council.

These data are all decision-making aids in the context of mobility plans, the development of commercial areas, bicycle plans or the installation of electrical terminals.

Artificial intelligence

Entropy founder Sami Kraiem was a researcher at the Vedecom Institute, which specializes in autonomous vehicles. “In 2015, artificial intelligence was emerging and we thought it could solve various problems relating to mobility. The purpose of this research and development project was to learn how the inhabitants move in order to then make predictions,” he recalls. The now entrepreneur had to train at HEC in order to pass the course of business. “Before, I had never sold anything. It was a new job,” says the young man with a smile.

Entropy is therefore an offshoot of the research institute Vedecom, “a spin-off” to use the vocabulary of start-ups. It employs seven people, and its goal is to double its workforce this year. “Using complex data, with a lot of information to manage, we need designers to simplify it and make it intelligible”, specifies Sami Kraiem.

Static and dynamic data

In concrete terms, three types of data, static and dynamic, submitted to artificial intelligence, make it possible to define a mobility scheme for each agglomeration, district by district. In open data, the start-up uses the population census (sex, age, number of cars per household, way of getting around, etc.) and the structural elements of a territory (places of employment, residence, shops, shopping centers, roads, etc.).

Then it buys data that specifically concerns mobility: GPS coordinates, train or bus tickets, subscription cards, etc. The algorithm combines all these elements and provides a fairly detailed description of travel.

This type of tool avoids the use of field surveys, but also the installation of sensors giving access to the same type of information. “Our ambition is that all territories can benefit from it. In interurban or rural areas, some communities do not yet know how the population moves,” says the entrepreneur, who intends to remedy this shortcoming.

Contact. Sami Kraiem, co-founder,

We wish to give thanks to the writer of this article for this remarkable material

To better organize trips, we predict them!

We have our social media pages here and additional related pages here.