AI & Mobility

The energy transition leads to new mobility concepts and vehicle electrification. Changing mobility patterns, e.g. by COVID-19, pose huge challenges. This SIG unites advances in operations research, data mining, and AI, using a human-in-the-loop approach. This is done in an interdisciplinary research cooperation including social, human-behavioral and geographical aspects. As such a multi-disciplinary AI approach is pivotal for some of societal biggest challenges.


Travellers will use combinations of different modes of public transportation and shared mobility more frequently. Because of its sustainable character, railway transportation is expected to play a crucial role in future transportation. Pressure on the railway infrastructure is foreseen, especially in urban areas. Consequently, management and control of future transportation pose huge challenges, where AI techniques can make a strong contribution. 

Public transportation

The Dutch transportation system is heavily loaded. Previous years, there was an enormous passenger growth. Currently, COVID-19 has an impact on the travel patterns. The system consists of many interlinked parts, making the system prone to disruptions that cascade easily through it. A combination of data-driven and algorithmic techniques helps to improve (integrated) operations management and disruption handling.

AI techniques are very promising to improve passenger information and interaction. This includes techniques to analyse massive passenger data sets, learning and optimisation strategies to manage passenger flows and find good travel advices, and application of human-centered AI techniques to influence passenger behavior and increase passenger satisfaction. A contribution of AI is foreseen to generate personalised multi-modal travel advices and the assessment of their benefits.

Shared mobility

The increasing importance of sustainability, the energy transition, and crowdedness of urban areas lead to different initiatives that may dictate the future of transportation via smart transport solutions such as  e-mobility, shared mobility (e-vehicles and bikes), Mobility-as-a-Service (MaaS) and also (Connected) Automated Vehicles. MaaS describes a shift away from personally-owned modes of transportation and towards mobility provided as a service, enabled by combining public and private transportation.

These new concepts require new algorithms and simulation for analysing journeys with different modes of transportation. This includes e-bike, train, and bus, but also walking and cycling. There also is a need new intelligent software applications that offer personalised, persuasive advices. Including psychological factors in mode of transport choice can strongly improve the prediction or description of human behavior and traffic flow. Besides effectiveness, efficiency, and user-satisfaction, social values such as safety, fairness, health aspects and environmental concerns need to be included.


We focus on applications within the above themes with the following challenges:

  • Data driven approaches: developing AI techniques including machine learning that process and interpret heterogeneous data from different sources to improve operations and influence behavior of travelers, where interpretability of results hasto be addressed.
  • Intelligent and hybrid planning algorithms: developing intelligent algorithms including optimisation methods and learning to solve planning and decision problems in the mobility domain.
  • Human-centered AI: developing AI techniques to understand and predict human choice behaviour and to persuade people to make better choices (efficient, environmentally friendly, etc.), including Intelligent interactive information presentations, AI-induced persuasive technology, personalised interaction to maximise user satisfaction.
  • Intelligent system simulation: develop AI techniques such as agent-based simuation to analyse the design and performance of mobility concepts, including decision models and roles of different actors.

Developing these AI approaches involves interdisciplinary research including:

  • Human factors and behavior: understanding human behavior while using AI. This relates to understanding human beliefs, needs, and desires, (in-)attention, decision making, and trust. The operators of AI systems need to be trained to interact with such systems more effectively and efficiently.
  • Governance of AI in mobility. This is crucial to ensure the public value that these systems will produce in terms of ethical norms and values, law and regulations, sustainability, safety, inclusion, and accessibility.