Open PhD position [TSE 133] - Predictive Resilience and Optimization Methods for Emergency Transportation Handling and Enhanced Urban Stability (PROMETHEUS) project

At the Chair of Transportation Systems Engineering (TSE) of the Technical University of Munich (TUM), we seek an excellent, motivated doctoral researcher for our TUM-funded research project. The research will be jointly supervised by Prof. Antoniou (TUM) and Prof. Stern at the University of Minnesota.

Description of the Chair

The Chair of Transportation Systems Engineering (TSE) undertakes research in the transportation field with a specific focus on modeling and simulating transportation systems, implementing data science and data analytics in transport and human factors analysis, and machine learning and deep learning. The TSE chair researches multimodal and unimodal freight and passenger transport demand and supply modeling, allowing for contributions to the optimization, calibration, and validation of transport models. In this direction, the application of big data acquisition and analysis and flexible data-driven models are examined. The TSE chair also contributes to analyzing human factors in transport-related fields such as road safety modeling, behavioral economics applications, and modeling of factors that affect user engagement in transportation systems.

Project description

Transportation systems provide essential services for the people who live in urban centres. However, these systems are not immune to disruption, and large-scale disruptive events, such as natural disasters and pandemics have the potential to substantially reduce the connectivity of communities. Moreover, the speed at which communities return to normal operation can be seen as a measure of a transportation system’s resilience. A key challenge in designing network-level control policies to allow transportation networks to quickly return to normal operating conditions is the ability to understand and model how the disrupted system will respond to different control inputs. In an ongoing collaboration, Prof. Antoniou and Prof. Stern have been studying how to use sparse nonlinear methods, such as SINDy to quickly model the disrupted system and implement control. In this work, we expand on this existing work to build a framework for rapid response to transportation network disruption. This will include two primary thrusts: (i) Disruption Detection and (ii) Disruption Mitigation. Together, these thrusts will provide methods to quickly identify transportation network disruptions through anomaly detection methods, and quickly model the disrupted system response, and finally close the loop by designing control schemes for the disrupted system dynamics to help the transportation network quickly return to normal operating conditions. The PhD position includes work on:

• Disruption measurement: The research team will develop different disruption metrics based on the type of mobility system being considered.
• Anomaly detection: Using the identified metrics, the research team will develop anomaly detection techniques to detect the presence of a disruptive event in the early stages.
• Reduced-order dynamical models for system behaviour: The research team will develop methods to learn a simplified representation of the network-level response dynamics to different control inputs.
• Optimization of response controller: Using the approximated surrogate dynamics, the research team will develop control techniques to adjust network capacity to minimize disruption.

The resulting work will provide valuable insights into resilient management of transportation networks under different types of disruptions. With increased severe weather events, as well as the global threat of terrorism and cyber-threats, these methods will help provide stability in an uncertain and highly connected world.

Prof. Dr. Constantinos Antoniou will supervise the PhD researcher at TUM together with Prof. Dr. Raphael Stern (University of Minnesota) who will spend some time in residence at TUM, and some time advising remotely from the University of Minnesota. Thus, the TUM PhD candidate will participate in regular joint meetings and formal workshops both virtually and in person.

Requirements

• Have an MSc degree in a relevant field (e.g., transportation engineering, data science, computer science);
• Be enthusiastic about researching transport-related projects and understand the fundamentals of transportation systems optimization and machine learning models;
• Have strong analytical skills;
• Have excellent research, academic writing, and presentation skills;
• Have a strong programming background and experience using e.g. Python or R;
• Have excellent working knowledge (written and oral) of English (knowledge of German will be considered a plus);
• Be able to work with strict deadlines.

In addition to the above, candidates must fulfil the general TUM admission requirements: https://www.gs.tum.de/en/gs/path-to-a-doctorate/application/requirements/

Conditions of Employment

Following the Public Sector Collective Agreement of Länder (TV-L), TUM offers a competitive compensation package. This position is a 75% TV-L 13. More information on the offered wages can be found at oeffentlicher-dienst.info/tv-l/west/

TUM is an equal-opportunity employer. Qualified women are particularly encouraged to apply. Applicants with disabilities are treated with preference, given comparable qualifications.