About
I am Luka Stärk, working on combinatorial optimization for railway and mobility applications. This page collects my projects, publications, talks, and other information.
Project
MobilityLab / Automated Railway Disposition Assistance.
Development of optimization algorithms for real-time disposition that supports control centers in finding wise dispatching decisions. Industrial partner: DB InfraGO AG with the automated dispatching assistant (ADA-PMB).
Publications
L. Stärk, R. Borndörfer, N. Lindner. From Routes to Resources: A Model for Real-Time Train Dispatching with No-Wait Constraints. ATMOS 2026, OASIcs vol. 147.
L. Stärk. TDP-NW: Benchmark Instances for the Train Dispatching Problem with No-Wait Constraints. Dataset, Zenodo, August 2026.
L. Stärk. Parameterized approximation algorithms for orienteering. M.Sc. Thesis, Technische Universität Berlin, 2023.
Awards
- 2025 1st Place DISPLIB 2025 Competition.
Talks
3 Sep 2026 From Routes to Resources - A Model for Real-Time Train Dispatching with No-Wait Constraints. ALGO/ATMOS 2026, L’Aquila.
17 Jul 2026 From Routes to Resources - A Model for Real-Time Train Dispatching with No-Wait Constraints. IFORS 2026, Vienna.
5 Mar 2026 A Parameterized Algorithm for Real-Time Train Dispatching. GOR AG LuV Workshop 2026, DXC Technology, Düsseldorf.
Short CV
2026 – PhD Student in Mathematics, Freie Universität Berlin, Institute of Mathematics; Berlin Mathematical School (BMS).
2025 – Research Associate, MODAL MobilityLab, Zuse Institute Berlin & Freie Universität Berlin.
2024 – 2025 Software Developer for Optimization, DB Systel GmbH (CTMS — Capacity & Traffic Management System).
2020 – 2023 M.Sc. in Computer Science, Technische Universität Berlin.
2020 – 2021 M.Sc. in Computer Science (ERASMUS), Université Grenoble Alpes.
Writing a Master’s Thesis at the MobilityLab
I co-supervise Master’s theses with Prof. Dr. Ralf Borndörfer (ZIB / FU Berlin) in combinatorial optimization, with a focus on railway and mobility applications. A solid background in algorithms, mathematical optimization, or theoretical computer science is expected.
Open topics:
- FPTAS, polynomial, incremental, and parameterized algorithms for soft-constraint shortest path problems in DAGs. Design and analyze efficient algorithms for resource-constrained shortest path variants on directed acyclic graphs. Each arc has a cost and duration. Objective nodes are given with a coefficient and threshold time. The objective is to minimize the total path cost plus the weighted tardiness at each objective node visited along the path.
If you are interested in one of these topics, have your own proposal, or are generally interested in finding a topic in optimization and mobility, send me an email with a short statement of interest.