Modeling transport management and land use over time

Abstract

Due to ever increasing travel demand, fiscal and environmental constraints, it is recognized that pure transport supply or pure demand management alone is not effective to mitigate traffic congestion. Developing integrated transport supply and demand management (TS–DM) strategies is crucial for ensuring sustainable urban development. TS–DM strategies will not only affect the transport system performance, but also induce changes in the land use pattern and hence changes in land value. Moreover, the implementation of TS–DM strategies typically involves a progressively phased schedule; one must account for the costs and effects that accrue over time. This paper develops a formulation to study the impact of TS–DM strategies on the overall system performance and activity location costs expressed as land value. Specifically, a nested multinomial logit model combined with the bid-rent process is formulated to model residents’ location and travel choices, with the problem of housing supply integrated in this framework. The overall combined network equilibrium problem is expressed as a non-linear complementarity problem. The existence and uniqueness of the equilibrium solutions are investigated through an equivalent mathematical programming formulation. Moreover, analytical results are derived to study the distribution of benefits due to transport infrastructure improvement among different stakeholders for networks with one origin–destination (OD) pair, for scenarios of homogenous and heterogeneous values of time. The analytical results show that transport improvements benefit landowners or developers rather than tenants under the scenario of homogeneous value of time; and benefit people with a higher income more under the scenario of heterogeneous value of time. Finally, a mathematical program is developed to determine the optimal TS–DM strategies over time in order to optimize the overall system performance. For general networks with multiple OD pairs, where analytical results are not available, a numerical example is provided to illustrate the effects of TS–DM strategies, which generally echo the analytical results developed for the case with one OD pair.