Breakthroughs made in the field of Artificial Intelligence, and in particular Deep Reinforcement Learning (DRL), bear the potential to optimize transportation systems in unprecedented ways. This includes railways, who need to boost capacity, quality, and efficiency to meet growing demand. Digitale Schiene Deutschland (DSD) and their AI partner InstaDeep are developing a central, DRL-based Capacity & Traffic Management System (CTMS) to automatically make timetables and real-time dispatching decisions for more than 40,000 train rides per day on the German rail network of over 33,000 km.

 

As product manager, Dr. Michael Küpper has led the development of the current CTMS prototype. He demonstrates, how the prototype controls a simulated regional rail system in near-real-time.

Dr. Waqaas Rehman, Business Development Manager at InstaDeep, highlights how only the amalgamation of deep domain expertise and cutting-edge AI capability reaps the full benefits of AI.