Special Issue
Materials - MDPI

Dr. Alfredo Núñez Vicencio – My Research

I consider railway infrastructure to be the backbone of any transportation system. But also, nowadays' green ambitions are only possible with a robust railway system. Still, the railway can contribute even more to society, and for that we still gotta face various challenges. My passion is the field of intelligent railway infrastructure. That is, equiping the railway with tools that allow the infrastructure to learn from previous experiences, to automatically assess its condition and maintain itself based on predictions from advanced intelligent models. In short, I develop the brains of a railway system.

Topics of my interest include health condition monitoring and maintenance decision support systems, with methodologies coming from computational intelligence, soft computing, machine learning, Big Data analytics, together with different non-centralized and centralized model predictive control approaches. I divide my research into two major subjects:

Monitoring and modeling of large-scale railway infrastructure systems

  • Converting intelligent health condition monitoring for railway infrastructures into useful information for maintenance purposes via information fusion methodologies.

  • Big data analytics and machine learning for railway monitoring data and other railway signals to extract crucial features for maintenance purposes from massive amount of railway data.

  • Design whole system key performance indicators to make possible the online evaluation and improvement of maintenance strategies for large-scale railway infrastructure.

  • Online large-scale railway monitoring and modeling, including the fusion of static, moving and crowd-based sensor monitoring technologies.

  • Innovation and implementation during operation.

Maintenance strategies for maintenance decision support of large-scale railway systems

  • Collaborative and non-collaborative game theory approaches, hierarchical and distributed model predictive control for maintenance decision coordinating different contract regions from a whole system perspective, minimizing misalignment between the objectives of contractors, ProRail, NS and railway users.

  • Hybrid model-based predictive control for railway systems including discrete and continuous variables, and different heuristics to solve expensive combinatorial problems. For instance, the design of maintenance strategies for rail-track replacement, grinding, tamping, are all large-scale mixed integer optimization problems.

  • Robust model predictive control for maintenance in railways, worst case analysis, scenario based approaches, optimistic optimization. Then, the effect of different sources of stochasticity can be systematically included in the control design.

  • Multi-objective model based predictive control (MO-MPC) for a dynamic support in maintenance decision processes in railway systems with multiple objectives. For instance, punctuality, efficiency, robustness, cost reduction, safety, sustainability and energy consumption of the railway operations.

Application domains:

  • Conventional railway tracks.

  • High speed lines.

  • Under-utilized rural/secondary "low density" lines.

  • Light trains (metro).

  • Freight dominated railway routes.