Reading time approx. 5 minutes
Accidents, extreme weather conditions, strikes or major events – when there is a problem in city traffic again, the dispatchers in the transport company control centers are faced with a truly mammoth task. Quick action is required to manage the chaos and keep public transport on track. But how can artificial intelligence help defuse these stressful situations and offer the dispatchers in the control centers valuable support?
This is exactly what is being worked on under the roof of the KARL Competence Center. The mission: To increase efficiency and responsiveness in local public transport through AI-supported systems. This is not just about technical solutions, but also about the human component – about how AI and human decision-makers can work hand in hand to make traffic in cities smoother.
Ariane Lindemann in conversation with Dr. Jochen Wendel, R&D Manager at INIT Group and project manager in KARL.
How can artificial intelligence relieve the workload of public transport dispatchers in stressful situations?
The goal of the project is to develop an AI model that supports dispatchers in the control centers. We want to develop an intelligent solution that allows operators to respond quickly and effectively to events such as accidents, construction sites, weather conditions or major events in order to maintain operations. By analyzing historical data and using AI algorithms, specific recommendations for action are generated for dispatchers. These recommendations range from rerouting vehicles and adjusting timetables to passenger information via various channels such as passenger displays or apps. These recommendations are provided in real time and can be integrated directly into the control center software. This reduces the burden on dispatchers and allows them to make informed decisions.
What role does real-time data play in the work of the AI model?
Real-time data is crucial to the performance of the AI model. By continuously monitoring data such as traffic volumes, weather conditions and service disruptions, the model can respond quickly and provide accurate recommendations. The integration of real-time data enables dynamic adaptation to changing conditions in public transport.
How to ensure that the AI model is reliable and user-friendly?
It is crucial that the AI model delivers reliable results and makes dispatchers feel like they are not being replaced. Transparency plays a big role here, as dispatchers need to understand how the model makes decisions and what data is taken into account. In addition, the results of the AI model must be presented clearly to ensure easy use.
What were the challenges in using data for the AI model?
A major hurdle was the preparation and anonymization of the data, as it was in different formats and contained sensitive information. Access to data, particularly personal data, proved particularly difficult. Thorough preparatory work was necessary to make the data usable. It had to be ensured that the historical data from the Intermodal Transport Control System, i.e. the control center software, was prepared and supplemented with further relevant information. It was also important to ensure the applicability and user-friendliness of the system by actively involving the dispatchers in the development process.
How do dispatchers react to the introduction of assistance systems?
Acceptance is generally high because assistance systems make work easier and, in particular, help new employees to learn the ropes more quickly. It is important to emphasize that the AI model serves as a support for the dispatchers and does not replace them – and that the decision-making power remains with humans.
What does the future of the project look like?
Although the project is still research in nature, we aim to integrate the developed solutions into future products. The challenges lie in both the technical implementation and the design of user-friendly tools that effectively support dispatchers. We are confident that the solutions developed in KARL can be put into practice in the future and will sustainably improve local public transport. The KARL project also has the potential to alleviate the impending shortage of skilled workers in public transport because new staff can be deployed more quickly.
To what extent could the project also be internationally relevant?
KARL has already attracted international attention. Particularly in North America, where the topic of automation in public transport is also of interest. The methods and solutions developed in KARL could also be useful in other countries with similar challenges such as weather events or traffic disruptions.
How are data protection and privacy taken into account when using data for the AI model?
Data protection and privacy are top priorities when using data for the AI model. All personal data is strictly anonymized and used exclusively for the development of the model. Strict security measures are implemented to control and monitor access to sensitive data.
What impact does the AI model have on the efficiency of public transport?
The AI model makes a significant contribution to the efficiency of local public transport by supporting dispatchers in quickly identifying and dealing with operational disruptions. By automatically analyzing data and providing relevant information, the model can help minimize outages and increase the reliability of public transport.
How are passengers’ needs taken into account when developing the AI model?
When developing the AI model, the needs of passengers are the focus. By optimizing public transport, journeys should become more reliable, comfortable and punctual. Passenger feedback is actively collected and incorporated into the further development of the model to ensure that it meets users’ needs and expectations.
About KARL
KARL’s goal is to design AI-supported work and learning systems in a human-centered, transparent and learning-conducive manner and to make them demonstrable in concrete practical applications. The project is aimed at companies, employees and interested parties in the Karlsruhe region who use AI-supported work and learning systems, deal with them or want to understand them better. KARL is one of currently 13 regional competence centers that deal with the effects of artificial intelligence (AI) on the world of learning and work.
The CyberForum is part of the project consortium and is primarily responsible for public relations, community management and the sustainability concept. The consortium leader is Karlsruhe University of Applied Sciences. In addition to seven research and transfer partners, the project partners also include ten regional companies.