Kentyou proposes innovative AI solutions by collaborating with visionary cities to help in their climate neutral transition, in particular, solutions designed for encouraging active and soft mobility.
In this experimentation, data from a large number of heterogeneous sources (demographic, topographic, points of interests, data about bike lane usage, traffic pattern, weather conditions, etc.) are integrated in the decision-making of the station allocation. Our data analysis process splits the area into geographical cells and by taking into account various parameters (e.g., social groups to incentivize, installation cost) as input to serve different decision policies of each municipality, the proposed optimization algorithm (based on p-median) decides whether to position a station in a cell or not.
Our goal is to validate our algorithm by using data and use case examples from EHubs4Data and to scale our solution by exploiting EUHubs4Data high performance tools and services (PSNC, TERALAB, EURECAT).
The main objectives of the experiments are:
- Strategically locate the soft mobility stations (shared/electric bikes, scooters, cars) in urban areas.
- Explore incentivization models for fostering the usage of soft mobility by citizens.
Our solution provides a unifying framework, able to easily integrate heterogeneous data into a unique processing environment. The unification provides adapters and wrappers for each relevant standard in order to include quickly and easily, at run-time, any data source of valuable information.
Our solution considers a wide variety of data (traffic, topographic), parameters (social groups) and municipality policies (cost) affecting the decision of the location-allocation optimization.
From an environmental perspective, our solution will reduce the use of individual cars, account for more than 16% of the greenhouse emissions in France, by encouraging bicycle and electric car usage.
And from a social perspective, our algorithm is specifically designed to ensure fairness in transportation access and to target different social groups (unemployed, ageing population). Furthermore, it promotes a healthier and more active lifestyle.
The Key Exploitable Result of this experiment are:
- Our main result will be a solution for multimodal mobility. The main value proposition will be to use local data sets to optimize the localization of mobility services such as bicycle sharing stations, Electric Vehicles charging stations, or other multimodal transport solutions.
- This will be proposed in a product + service approach, as each local community has different available datasets, but also different requirements in term of localization and priorities.
KPI 1 - Integrate new datasets in sensiNact platform and analyze their impact on the localization problem.
KPI 2 – Decrease the computational performance of our optimization algorithm by an order of magnitude. (Initial value > 1 h)
KPI 3 – Develop an application with policy reconfiguration features for the municipality, including incentivization techniques for soft mobility.
KPI 4 – Increase the maturity of our solution to enable a prototype demonstration (initial TRL = 5)
KPI 5 – Formalize a business model for the KER and integrate it in the company’s business strategy.