The project EMERALDS, where R&D Spain led the task Mobility AI-as-a-Service, ended successfully on the 31st of December 2025 after 3 years of work involving the MLOps team.

At the final review held on March 4 in Athens, the consortium presented the project's outcomes, consisting of a set of EMERALDS services for mobility data processing and extreme-scale mobility data analytics and learning, together with their application and validation on three use cases: UC1 - Risk Assessment, Prediction and Forecasting in crowded events; UC2 - Multi-modal traffic network management and UC3 - Trip Characteristics Inference and Traffic Flow Data Analytics, as well as early adoption demonstrators.

Some of the main results have been the following:

• EMERALDS Service on Mobility Data Fusion and Management
• EMERALDS Service on Extreme Scale Mobility Data Analytics at the Edge/Fog/Cloud continuum
• EMERALDS Service on Active & Federated Learning over Mobility Data
• Development and validation of the Mobility AI-as-a-Service platform (Developed by BULL)
• During the Final Review, a video demonstration of the Mobility AI as a Service Platform was showcased.
   

R&D Spain's contribution has been key to the project because the Mobility AI as a Service (MAIaaS) platform developed was designed to manage the lifecycle of more than 15 AI/ML models used within the EMERALDS services developed in the project. The Mobility AI as Service was the source of intelligence within the project architecture to develop smart mobility services.

The MAIaaS platform design is totally aligned with our MLOps strategy and contributed to enhance different modules of our MLOps framework. In particular, the MAIaaS platform introduced relevant innovations related to our Model Monitoring Module, where monitoring capabilities were introduced exploiting Evidently AI and Grafana.