共享停车泊位数量可变下交通流逐日演化规律

为了揭示在共享停车泊位数量可变条件下的网络交通流逐日演化规律,首先构建了共享泊位交易系统,并考虑了交易市场中的共享泊位提供者可以选择2种异质性的价格预期方式,即理性预期方式和幼稚预期方式;而后对共享泊位的均衡价格、2种提供者的占比差、高峰时段公交和小汽车需求的演化规律进行了分析;其次,以2条平行路径的路网为例,对网络交通流量分配的最终演化结果进行了分析;最后,在对上述2个系统的最终演化状态给出定量判据后,以北京市实际路网为例进行了数值试验。理论分析和数值试验结果表明:①对于共享泊位交易系统,若供给曲线斜率小于需求曲线斜率,则共享泊位交易系统的唯一均衡解可实现无条件渐进稳定;否则若理性提供者的交易与预测成本之和大于幼稚提供者,则存在临界提供者选择强度,使得共享泊位交易系统在大于此临界值条件下出现分岔或混沌现象;②对于网络交通流系统,若出行成本对路径流量敏感度小,路径选择概率对出行成本敏感度小,小汽车需求量不大,则系统唯一的均衡解可能是渐进稳定的,否则系统会出现分岔或混沌状态;③当共享泊位交易系统处于渐进稳定状态时,若提供者对共享泊位的价格变动不敏感,用户对其价格变动敏感,潜在共享泊位需求量不大,理性提供者的交易与预测成本之和并非远大于幼稚提供者,提供者选择强度不大,则由于受到共享泊位交易总量的限制,高峰时段的均衡小汽车需求不大,导致网络交通流系统的最终演化状态容易趋向于渐进稳定;④当共享泊位交易系统处于混沌状态时,网络交通流系统会产生更加严重的分岔与混沌现象。

Day-to-day Dynamical Evolution of Traffic Flow with Variable Quantity of Shared Parking Lots

This study aimed to reveal the evolution rule of day-to-day traffic flow with a variable quantity of shared parking lots.To achieve this,a trading system for shared parking lots was developed.Shared parking providers can choose between rational and naive expectations of the prices of shared parking lots in the trading system.On this basis,we investigated the evolution rule of the equilibrium price of shared parking lots,proportion of two types of parking providers,and travel demands of public transit and private vehicles during rush hour.Subsequently,taking a network with two parallel roadways as an example,we examined the final state of the dynamical evolution of the network flow system.A quantitative criterion is provided for the final evolution state of both systems.A numerical experiment with a real traffic network in Beijing was conducted.Theoretical analysis and numerical results lead to the following conclusions:① Concerning the trading system of shared parking lots,if the slope of the supply curve is less than that of the demand curve,the unique equilibrium solution of the trading system of shared parking lots can always achieve an asymptotically stable state.If the summation of transaction and forecasting costs for rational providers is higher than that for naive providers,a critical choice intensity of parking providers exists,which may lead to bifurcation and chaos in the trading system.② For a network flow system,the unique equilibrium solution may be asymptotically stable if the travel cost is not sensitive to the path flow.The selected path probability is not sensitive to the travel cost if the vehicle demand is not large;otherwise,the system can incur bifurcation and chaos.③ When the trading system reaches an asymptotically stable state,if parking providers (users) are insensitive (sensitive) to parking prices while the potential parking demand is low,the summation of transaction and forecasting costs for the rational provider is not much higher than that for the naive provider,and the choice intensity of parking providers is not large.The equilibrium demand for private vehicles during rush hour is not high because of the limited number of reserved shared parking lots,which leads to a final evolutionary state of the network flow system tending to be asymptotically stable.④ Once chaos occurs in the trading system of shared parking lots,the network flow system may generate more serious bifurcation and chaos.