Florian Nachtigall | Technical University of Berlin & Mercator Research Institute on Global Commons and Climate Change

About Me

I am a PhD candidate at the Technical University of Berlin. My research focuses on how machine learning techniques can advance and scale-up the decarbonization of the transport and building sectors. Methodologically, I’m interested in Graph Neural Networks (GNNs) for modeling spatial relationships and causal machine learning for estimating the effects of transport policies and infrastructure interventions on travel behavior and related emissions.

I am supervised by Prof. Dr. Felix Creutzig and affiliated with the Mercator Research Institute for Global Commons and Climate Change (MCC), where we aim to support data-driven solutions for low-carbon urban planning using geospatial data and machine learning as part of the AI4UP project.

News

[Jul. 2024] Tutorial on predicting mobility demand from urban features for the Climate Change AI summer school (see tutorial & code)
[Jun. 2024] Presentation on evaluating low-carbon urban planning strategies using causal machine learning and identifying housing locations that minimize travel (see slides) (WhatWorks Climate Solutions Summit - Conference)
[Dec. 2023] Presentation on using double machine learning to disentangle the non-linear influence of the built environment on travel behavior from residential self-selection (see recording, poster & short paper) (NeurIPS - Workshop)
[Sep. 2023] Presentation of the gtfs2nx Python package and its application for equitable and sustainable transit accessibility analysis (see slides) (GIScience - Workshop)
[May. 2023] Presentation on spatially-explicit causal analysis of urban densification and infrastructure interventions (Machine Learning in Economics - Workshop)

Key Publications

NeurIPS Workshop

The built environment and induced transport CO2 emissions: A double machine learning approach to account for residential self-selection

Florian Nachtigall, Felix Wagner, Peter Berrill, Felix Creutzig

NeurIPS Workshop, Tackling Climate Change with Machine Learning, 2023.

PDF Poster

CEUS

Predicting building age from urban form at large scale

Florian Nachtigall, Nikola Milojevic-Dupont, Felix Wagner, Felix Creutzig

Computers, Environment and Urban Systems, 2023.

PDF Code

Sci. Data

EUBUCCO v0.1: European building stock characteristics in a common and open database for 200+ million individual buildings

Nikola Milojevic-Dupont, Felix Wagner, Florian Nachtigall, Jiawei Hu, Geza Boi Brüser, Marius Zumwald, Filip Biljecki, Niko Heeren, Lynn H. Kaack, Peter-Paul Pichler, Felix Creutzig

Nature Scientific Data, 2023.

PDF Code Data Project Page

Other Publications

npj

Towards a public policy of cities and human settlements in the 21st century

Felix Creutzig, Sophia Becker, Peter Berrill, Constanze Bongs, Alexandra Bussler, Ben Cave, Sara M. Constantino, Marcus Grant, Niko Heeren, Eva Heinen, Marie Josefine Hintz, Timothee Ingen-Housz, Eric Johnson, Nina Kolleck, Charlotte Liotta, Sylvia Lorek, Giulio Mattioli, Leila Niamir, Timon McPhearson, Nikola Milojevic-Dupont, Florian Nachtigall, Kai Nagel, Henriette Närger, Minal Pathak, Paola Perrin de Brichambaut, Diana Reckien, Lucia A Reisch, Aromar Revi, Fabian Schuppert, Andrew Sudmant, Felix Wagner, Janina Walkenhorst, Elke Weber, Michael Wilmes, Charlie Wilson, Aicha Zekar

npj Urban Sustainability, 2024.

PDF

Nature CC

Demand-side strategies key for mitigating material impacts of energy transitions

Felix Creutzig, Sofia G Simoes, Sina Leipold, Peter Berrill, Isabel Azevedo, Oreane Edelenbosch, Tomer Fishman, Helmut Haberl, Edgar Hertwich, Volker Krey, Ana Teresa Lima, Tamar Makov, Alessio Mastrucci, Nikola Milojevic-Dupont, Florian Nachtigall, Stefan Pauliuk, Mafalda Silva, Elena Verdolini, Detlef van Vuuren, Felix Wagner, Dominik Wiedenhofer, Charlie Wilson

Nature Climate Change, 2024.

TR-D

Comparing urban form influences on travel distance, car ownership, and mode choice

Peter Berrill, Florian Nachtigall, Aneeque Javaid, Nikola Milojevic-Dupont, Felix Wagner, Felix Creutzig

Transportation Research Part D: Transport and Environment, 2024.

PDF

Open Source Software

gtfs2nx: A Python package to convert GTFS feeds to realistic, routable NetworkX graph.
slurm-pipeline: A thin Python wrapper for resilient and scalable Slurm job scheduling.

Teaching

[WS 2022/23] Teaching assistant for the master course Deep Learning at the Hertie School