An efficient approach to solving the road network equilibrium traffic assignment problem.

This paper presents a solution technique for large scale road network equilibrium assignment and related flow problems with nonlinear costs. It is shown that this nonlinear network problem can be solved without explicitly considering any of the constraints—they are satisfied automatically in the procedure developed—and without storing all of the individual decision variables. The computational requirements of the algorithm reported are almost identical to the requirements of heuristic solution techniques for traffic assignment. A solution time of nine seconds on the CDC 6400 was obtained for the test problem, which included 76 arcs and 24 nodes, all of which were origins and destinations.