Pandana
Pandana is a Python library for network analysis that uses contraction hierarchies to calculate super-fast travel accessibility metrics and shortest paths. The numerical code is in C++.
v0.8, released August 24, 2026.
Origins and acknowledgments
Pandana originated in research led by Paul Waddell at the University of California, Berkeley. Waddell conceived the network-accessibility application with Fletcher Foti, then his doctoral student and graduate student researcher, and Dennis Luxen of the Karlsruhe Institute of Technology, and directed and funded the initial development.
Foti led implementation of the original software and its pandas-oriented interface. Luxen contributed contraction-hierarchy expertise and code developed through the Open Source Routing Machine project. Initial development was supported in part by National Science Foundation award IIS-0964412 and by the Metropolitan Transportation Commission.
Subsequent development has included major contributions from Matt Davis, Federico Fernandez, Sam Maurer, Joaquim Gromicho, Eli Knaap, and other community contributors. The repository’s HISTORY.md file provides additional context, while the Git history remains the authoritative record of individual code contributions.
A paper on Pandana was presented at the Transportation Research Board Annual Conference in 2012. Please cite this paper when referring to the methodology implemented by this library.