Multimodal Network Models for Robust Transportation Systems

摘要

Since transportation infrastructure projects have a lifetime of many decades, project developers must consider not only the current demand for the project but also the future demand. Future demand is of course uncertain and should be treated as such during project design. Previous research for Southwest Region University Transportation Center (Report 167556) explored the impact of uncertainty on roadway improvements and found neglecting uncertainty to lead to suboptimal network design decisions. This research is extended in the current work by considering not only motor vehicle traffic, but other modes as well. The first half of this report examines the problem of network flexibility in the face of uncertainty when constructing a potentially revenue-generating toll road project. Demand uncertainty and network design are considered by way of a bilevel stochastic recourse model. The results from a test network, for which a closed form solution is possible, indicate that the value of network flexibility directly depends on initial network conditions, variance in future travel demand, and toll pricing decisions. The second half of this report integrates Environmental Justice into the transit frequency-setting problem while considering uncertainty in travel demand from protected populations. The overarching purpose is to improve access via transit to basic amenities to: 1) reduce the disproportionate burden transit dependent populations experience; and 2) increase the financial security of low-income households by giving them a feasible option to reduce their dependence on autos. The example application illustrates the formulation successfully increases access to employment opportunities for residents in areas with the high percentages of low-income persons, as well as demonstrates the importance of considering uncertainty in the locations of low-income populations and employment