01127505

Component

This study examines the use of severe capacity-disruptions in lieu of link-removal, in assessing transportation network robustness in the presence of isolated sub-networks. A procedure that utilizes capacity-disruption will be immune to the effects of poor connectivity and isolating links in the network being studied. One of the goals here was to build on the recently proposed NRI measure to find the best capacity-disruption level to use by analyzing the NRI values for each of 3 hypothetical networks with varying levels of connectivity. A new form of the metric, NTR, provides scalability for the network robustness measure. The primary conclusion from this investigation is that the use of complete link-removal to model network robustness is not only infeasible for networks with isolating links, but also does not yield unique results due to the influence of Braess’ Paradox involved with this computational procedure. Analysis of the NTR values suggests an upper limit of 99% on disruption levels to be used in robustness analysis. However, this upper limit may fall as low as 95%, depending on network connectivity. The rank-order analysis indicates that the most stable range for the rank-orders will vary with the level of connectivity of the network, but is likely to fall between 75% and 99%.

01120148

09-2644

English

2009

Transportation Research Board 88th Annual Meeting

DVD

Figures (3) ; References (38) ; Tables (3)

Connectivity; Highway capacity; Network links; Public transit; Systems; Systems analysis; Transit operating agencies

Network Robustness Index

Road networks

Highways; Planning and Forecasting; Public Transportation; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2009 Paper #09-2644

TRIS, TRB

Jan 30 2009 7:00PM