Computing and Analyzing Equilibrium Network Flows of Gasoline and Electric Vehicles

摘要: This paper addresses a new traffic assignment problem with mode and route choices for the emerging need of modeling networks that accommodate electric vehicles. Two transportation modes (or vehicle types), gasoline vehicles, are exclusively included in problem, which distinguish from each other terms driving distance limit travel cost composition. A convex programming model is proposed as tool evaluating such mixed-mode anticipated to exist long period electrified era. The focus this on computation evaluation problem’s solutions. In particular, authors developed implemented two competitive solution algorithms: one linear approximation algorithm Frank-Wolfe type, allows parallel treatment O-D pairs one-to-all constrained shortest path procedure generation; another quadratic makes use Gauss-Seidel decomposition deal sequential manner generate paths by one-to-one procedure. Experimental results applying these algorithms number synthetic realistic clearly show that, behavior perspective, produced mode-route flows replicate choice pattern, and, efficiency relative competitiveness depends required precision.