Investigating Boundary Issues Arising from Congestion Charging in a Bottleneck Scenario

Step tolling with bottleneck queuing congestion

“…To numerically solve his model, Chu uses a cobweb procedure. Ge and Stewart (2010) use the cell transmission model which combines flow-congestion on each link and queuing upstream when the inflow rate exceeds link capacity. They consider several toll schedules including step tolls (with a rapid, but finite, rate of increase and decrease in the toll to maintain continuity).…”

Dynamic traffic modelling for travel demand management

“…Different from the design of congestion pricing, this paper investigates derived issues after the implementation of congestion pricing; hence DUE is used in this paper and congestion pricing cost is part of travellers' generalised travel cost. The issue of boundary effects has been investigated in May and Milne (2000), Ge and Stewart (2010), Stewart and Ge (2014), Ge et al (2015b). This investigation attributes the undesired spatial and temporal boundary effects to improperly designed congestion tolls in road networks and it is concluded that well-designed time-varying tolls can significantly reduce the undesired boundary effects.…”

Investigating undesired spatial and temporal boundary effects of congestion charging

“…The boundary issues of interest are the fact that the implementation of a congestion charging project may drive too many travellers to travel outside the charging period so that undesired congestion occurs just before or after the period. Ge and Stewart (2010) investigates the temporal boundary effect in a bottleneck scenario, which uses a more refined bottleneck model than that of Vickrey's. In the Vickrey model (Vickrey 1969), traffic congestion takes the form of cars queuing behind a bottleneck without occupying any physical space, which is recognised to be a key limitation of these types of bottleneck models.…”

“…In the Vickrey model (Vickrey 1969), traffic congestion takes the form of cars queuing behind a bottleneck without occupying any physical space, which is recognised to be a key limitation of these types of bottleneck models. Ge and Stewart (2010) instead treats a bottleneck as a real road segment or link with a limited capacity, whose upstream link has a higher capacity; hence the vehicles in the bottleneck can queue up into the upstream link. Traffic is assumed to propagate through the bottleneck following the kinematic wave (KW) model of traffic flow.…”

“…It is these settings that make the new bottleneck model readily produce temporal boundary issues. The investigation in Ge and Stewart (2010) shows that the choice of congestion charging period(s), constant toll levels and the rate of change in tolls over time all exert influence over boundary issues.…”