Understanding route choice behaviors' impact on traffic throughput in a dynamic transportation network

Route planning is one of the most important and attractive topics in complex networks, geographical information science (GIS) and logistics. Travelers' route choice behaviors may affect the actual throughput of a transportation network. This study aims to analyze the influence of travelers' route choice behaviors on network traffic throughput. Two kinds of route choice behaviors, called continuous optimization behavior and hybrid optimization behavior, are defined. Based on this, two traffic scenarios are constructed to describe these behaviors and examine their effect on the network's actual traffic throughput. The application of our approach to urban street networks, Barabási-Albert scale-free network and Erdös-Rényi random network demonstrates that if all travelers choose optimal paths constraint with time from the start, network throughput is insensitive to the continuous optimization behavior but depends on the relationship between time and distance in the routing algorithm. Besides, network throughput is sensitive to the hybrid optimization behavior that the more travelers choose time-dependent optimal paths, the higher the throughput. The results indicate that the choice of initial paths and the time dependence of routing algorithms can directly affect the network's actual throughput which is enhanced >50 % when the time dependence parameter is >0.2 or the proportion of time-dependent shortest paths in initial paths is >0.2. Our findings provide supports for exploring efficient routing algorithms to enhance the throughput of urban street networks and other kinds of networks.