Solving the Dynamic User Equilibrium Problem Via Sequential Convex Optimization for Parallel Horizontal Queuing Networks

摘要: This article considers the dynamic user equilibrium (DUE) problem for parallel networks. The network dynamics are modeled using a Godunov discretization of Lighthill-Williams-Richards partial differential equation with trapezoidal flux function. model is augmented an additional constraint that prevents vehicle holding which flaw in discretization. departure rates assumed to be fixed. Under these assumptions, authors show future allocation demand among different paths at origin has no effect on travel time vehicles already network. enables them DUE fixed steps horizon can decomposed into series static UE problems and solved sequentially. Thus, as sequence convex optimization problems.