Optimising public transport with AI in Bergen

Phase 1 – Completed

In February 2021, Bergen Light Rail planned extensions of the light rail system in the city – from a single line of 20 km to a complete network. Our task was to propose an expansion solution. The objectives were to make future transport scenarios for light rail and suggest infrastructure to optimise both operational and capital expenditure.

Phase 2 – In progress

Vestland County Council wants a predictive model for Bergen’s entire public transport (PT) network to connect the light rail system to the bus network. The objectives were to propose optimisation for the overall travel patterns and get an early indicator of the consequences that new developments can have for mobility in the city.

Asistobe is the first to introduce an affordable SaaS tool that re-invents the planning and execution of PT and multimodal system. It does this by:

• Combining data sources to get deep understanding of how people move
• Use GDPR-compliant telecom data to get an accurate model for real transport demand (used in Phase 2)
• Advanced maths, statistical methods, and queuing theory
• AI to improve data and look for short- and long-term trends/patterns

The tool enables planners to make smart decisions by:

• Exploring how people actually move in the city
• Predicting future mobility and building detailed PT scenarios based on real transport demand and urban plans
• Optimising PT given the chosen scenario and measuring continuously the effect of the plans put into operation

Phase 1

Compared to the Light Rail without optimisation, the tool proposed operational patterns that reduce the kilometres driven, while increasing the capacity in high-density areas, all without the need of purchasing additional rolling stock and sacrificing traveller comfort levels. The impact in numbers is:

• Operational expenditure savings of 23%
• Yearly passengers increased 25%
• CO2 emissions reduced 30%
• As a result, €22m in annual value creation

Phase 2

Since the project is in progress, there are no results yet. However, the early analysis makes us expect around a 10-15% reduction in operational expenditure for the whole public transport system in Bergen, hence potentially making significant improvements for people in Bergen and the city’s carbon footprint.