Generalized Markov Chain Modelling for Networks¶
MCRoute is a Python package that enables the creation, manipulation, and study of networks in which movement along networks is characterized by a Makrov chain process.
MCRoute extends (and restricts) the NetworkX module to require that nodes and edges have associated transition probability matrices. Networks consist of directional graphs (NetworkX’s DiGraphs).
With MCRoute you can rapidly prototype Markov chain networks and transition probability matrices using theoretical distributions or by sampling directly from data.
Audience¶
This module was originally built to support a theoretical definition and research paper, but is intended to allow both researchers and practitioners, specifically in the field of transportation, to create and analyze Markov chain networks. Initial features were deisgned to support specific use cases, generally around schedule adherence and path reliability on a transit network. This may change as applications are expanded.
If there is a particular analysis you would like to do that requires a specific feature not yet implemented you can either suggest a change or attempt an implementation yourself.
Free Software¶
MCRoute is free software; you can redistribute it under the terms of the GNU General Public License, version 3, provided that you extend the same rights to those using your software. We welcome contributors, join us on GitHub