Berlin's BVG uses quantum computing to fix its driver shortage crisis

Berlin’s public transport operator, BVG, has turned to quantum computing to tackle a major scheduling challenge. The company faces a growing workforce crisis, with over 4,300 employees set to retire by 2026. A new quantum-based solution has now optimised schedules for 150 drivers across multiple bus lines, offering a potential lifeline for the struggling system. The project focused on a long-standing issue: classical scheduling methods often ignore driver preferences, leading to inefficiencies. By using Kipu Quantum’s Bias-Field DCQO algorithm, the team integrated individual preferences into the process. They also cut API calls by 80% through DBSCAN clustering, streamlining the system.

An Uncertainty Adapter played a key role, combining an Isolation Forest anomaly detector with a Gaussian Process demand predictor. This hybrid approach ensured the solution could handle real-world unpredictability. The result was a 2% efficiency gain—a small percentage that, at BVG’s scale, translates to roughly €18 million in annual savings. Beyond immediate benefits, the project proved that a classical-quantum-classical pipeline could move from early-stage testing (TRL 4) to a production-ready pilot (TRL 6) in just 24 months. The architecture’s flexibility means it could also apply to other complex logistical problems, such as hospital shift planning or last-mile deliveries. The breakthrough comes as BVG grapples with staff shortages, worsened by ongoing voluntary resignations on top of the looming retirements. Efficient scheduling could help ease pressure on the remaining workforce while maintaining service levels.

The quantum-based system has already shown measurable improvements in scheduling efficiency and cost reduction. With BVG’s workforce under strain, the technology offers a practical way to manage resources more effectively. The success also opens doors for similar solutions in other industries facing complex logistical demands.