Skip to content

Codes in this repository compute the traffic assignment using the algorithms of: (1) Frank-Wolfe Algorithm with Exact Line Search/step size of (2 / (k + 2)); (2) Conjugate Direction Frank-Wolfe Algorithm; (3) Path-based Projection Gradient Algorithm with Exact Line Search/fixed step size

yuzhenfeng2002/Traffic-Assignment-2024

 
 

Repository files navigation

Traffic-Assignment-2024

Codes in this repository compute the traffic assignment using the algorithms of:

  • Frank-Wolfe Algorithm with Exact Line Search
  • Frank-Wolfe Algorithm with step size of (2 / (k + 2))
  • Conjugate Direction Frank-Wolfe Algorithm
  • Path-based Projection Gradient Algorithm with Exact Line Search
  • Path-based Projection Gradient Algorithm with fixed step size

You can see the performance:

You can see the references:

  • Jayakrishnan R, Tsai WT, Prashker JN, Rajadhyaksha S, 1994 A faster path-based algorithm for traffic assignment. Transportation Research Record .
  • LeBlanc LJ, Morlok EK, Pierskalla WP, 1975 An efficient approach to solving the road network equilibrium traffic assignment problem. Transportation research 9(5):309–318.
  • Mitradjieva M, Lindberg PO, 2013 The stiff is moving—conjugate direction frank-wolfe methods with applications to traffic assignment. Transportation Science 47(2):280–293.

Acknowledgments

This work is based on Traffic-Assignment-Frank-Wolfe-2021. I focused on fixing this implementation and extending it to implement more algorithms.

About

Codes in this repository compute the traffic assignment using the algorithms of: (1) Frank-Wolfe Algorithm with Exact Line Search/step size of (2 / (k + 2)); (2) Conjugate Direction Frank-Wolfe Algorithm; (3) Path-based Projection Gradient Algorithm with Exact Line Search/fixed step size

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%