AI support in real time

Intelligent control system for traffic prioritization

A major traffic intersection required an intelligent control system to optimize the prioritization of mobile units. The goal was to enhance average speed and prevent delays. By implementing artificial intelligence (AI), the process was successfully improved.

Close-up of screen showing software code

Challenge

Manual prioritization of mobile units at a traffic intersection led to delays. The reason: The optimization potential in the control system remained untapped.

Consulting approach

By integrating data pipelines and training an AI using reinforcement learning, an intelligent agent was created to provide crucial support.

Client benefits and solution

The intelligent control system assists the dispatcher in real time. Delays are significantly reduced through optimization of the average speed.

locomotive engineers in discussion beside locomotive

The Challenge in Detail:

In the past, the prioritization of mobile units passing through a traffic intersection was done manually. Optimizing the average speed was not feasible, often leading to delays. Due to the high complexity and dynamic nature of the situation, an analytical solution was not effective – that’s why valantic focused on an AI-based solution. Simulations and historical data served as the training foundation for the system.

Solution & Results in Detail:

The AI project was realized through the following measures:

Engineering of data pipelines from various IT systems to enable simulations for training a machine learning model
Development and training of an intelligent agent using reinforcement learning: The agent learns to make optimal prioritization decisions based on the available data.
Implementation of a minimum viable product (MVP) into IT systems to support dispatchers in determining the sequence of mobile units in real time

Your Contact

Laurenz Kirchner

Partner & Managing Director

valantic