Enhancing Urban Road Network by Combining Route Planning and Dynamic Lane Reversal

Traffic jam in the urban area has been a global problem for a long time. Previous methods either adopt the route planning (RP) policy to bypass congested areas, or use the dynamic lane reversal (DLR) policy to temporarily increase road capacity in one direction. To further enhance urban road network, in this paper, we propose to combine RP and DLR to save vehicle’s travel time. DLR with a relatively long period locally improves the road capacity in the desired direction, while RP with a relatively short period computes optimal routes in the global area. To evaluate the proposed method, we use the simulator SUMO to build 2 kinds of road networks, a small-scale grid network and a large-scale OSM network, and on this basis, tune the parameters of the proposed method. Extensive experiments confirm that the proposed method reduces the average travel time by 84.4%, 74.6% and 81.1% respectively, compared with the basic method (without any control), DLR, and RP, in the large-scale OSM network. Meanwhile, the CO
2
emission and fuel consumption are greatly reduced as well.