Robustness, efficiency and recovery of public transport systems

Project overview

The project developed operations research methods to increase the efficiency, reliability and resilience of urban and interurban public transport systems in contexts of competition and cooperation.

Its research covered strategic, tactical and operational planning, considering infrastructure design, service planning, passenger demand and the ability of transport networks to recover from disruptions.

Research focus

The project combined mathematical programming, stochastic optimisation, decomposition methods, big data and machine learning. The work addressed:

1. robust and recoverable design of public transport networks and services;
2. distributed surrogate optimisation for computationally expensive problems;
3. airline and high-speed railway planning under competition and uncertainty;
4. operational adaptation to demand variability and dynamic demand estimation;
5. machine learning methods for transport demand modelling;
6. robust airport and airspace design and disruption recovery.

The project also explored dynamic passenger-demand estimation using Kalman filtering and discrete-choice models based on machine learning.

Contribution

The research established methodological foundations for integrating robustness, recoverability and data-driven demand modelling into public transport planning. It also supported the development of software, doctoral research and scientific publications in transport optimisation and machine learning.