May 13th 2024
AI cases for public transport
Artificial intelligence (AI) can support public transportation in many areas. At IT-Trans, Lufthansa Industry Solutions (LHIND) will be presenting specific examples of AI applications. From the creation of annual sales forecasts, which deliver reliable results even in times of crisis, to AI support for transportation planning.
In their presentation "How to empower your business with AI use cases", Dr. Wilfried Bohlken, AI Specialist at LHIND, and Business Manager Jascha Koepsel will present use cases of how AI can help in public transport at the Market Update Forum on 15 May from 4 to 5 pm:
Looking to the future: better control with AI
Many transport companies use simple formulas or statistical methods to calculate annual revenue forecasts. "If unforeseen events occur, like the coronavirus pandemic, such static models quickly reach their limits," says Dr. Wilfried Bohlken. "AI, on the other hand, recognizes correlations between time series data and external factors - such as current population figures, the number of cars in a supply area or vacation periods - and uses these for forecasting." This makes it possible to explain cyclical patterns and linear trends in the available data, explains Bohlken.
How do short-term changes in demand affect sales? During the coronavirus pandemic, LHIND worked with a customer to create an AI forecasting model (blue) that was able to predict the real values (black) very accurately over a 12-month period.
AI in transportation planning: passenger numbers, detour and computer vision
Passenger numbers are central to transportation planning. However, due to numerous practical challenges, they can rarely be reliably determined in retrospect. AI can provide a solution: Using historical data, an algorithm projects passenger numbers for existing gaps. Data from the last 12 months can be automatically extrapolated retrospectively for individual days and stops. "Our machine learning routines ensure that the algorithm is constantly learning and incorporates it into the monthly recalculations. This enables demand-oriented capacity planning," says Bohlken.
AI can also access map services and automatically compile information on detours. Drivers can view the latest operating instructions directly on their smartphones or tablets, and differences in mileage are automatically recorded and forwarded. Cameras that are already in use on platforms can also become smarter thanks to AI: Thanks to computer vision, the system provides real-time information on how many passengers are on the platform or whether there are people on the track bed. This means it can always respond to current requirements.
The experts from LHIND will also be showcasing these and other examples at IT-Trans in Hall 1, Stand B40. There, they will also provide insights into the latest developments of the Patris sales background system, which will also be equipped with AI functions in the future and will be available as an app.