Modeling an Elastic-Demand Bimodal Transport Network with Park-and-Ride Trips

This paper presents a network equilibrium formulation for modeling commuters＇ travel choices in a bimodal transport system with park-and-ride （P＆R） trips while the total demand is elastic to the congestion level of the network. A super-network approach is adopted in the proposed model. It is assumed that commuters＇ trips are categorized into two types, auto mode only and a combined mode with both auto and transit modes. The former is referred to as the pure mode trip and the latter as the P＆R mode trip. The proposed model simultaneously considers the commuter＇s choice of the pure mode versus the P＆R mode, the choice of parking location for the pure mode, the choice of transfer point for the P＆R mode, as well as the route choice for each mode. The demand elasticity of transport system, the capacity constraints of transport facilities, and the congestion interaction throughout the super-network are also explicitly incorporated into the proposed model. The results of the numerical experiment show the following key findings： （i） traditional parking/P＆R models may overestimate or underestimate travel demand distribution over network; （ii） parking/P＆R, transit scheduling, and carpooling schemes bring significant impacts on commuters＇ travel behavior and network performance; and （iii） different transport policies may be to some extent mutually substituted .

Modeling an Elastic-Demand Bimodal Transport Network with Park-and-Ride Trips

This paper presents a network equilibrium formulation for modeling commuters' travel choices in a bimodal transport system with park-and-ride (P&R) trips while the total demand is elastic to the congestion level of the network. A super-network approach is adopted in the proposed model. It is assumed that commuters' trips are categorized into two types, auto mode only and a combined mode with both auto and transit modes. The former is referred to as the pure mode trip and the latter as the P&R mode trip. The proposed model simultaneously considers the commuter's choice of the pure mode versus the P&R mode, the choice of parking location for the pure mode, the choice of transfer point for the P&R mode, as well as the route choice for each mode. The demand elasticity of transport system, the capacity constraints of transport facilities, and the congestion interaction throughout the super-network are also explicitly incorporated into the proposed model. The results of the numerical experiment show the following key findings: (i) traditional parking/P&R models may overestimate or underestimate travel demand distribution over network; (ii) parking/P&R, transit scheduling, and carpooling schemes bring significant impacts on commuters' travel behavior and network performance; and (iii) different transport policies may be to some extent mutually substituted.