Link Disruption Scenario Generation for Transportation Network Criticality Analysis

Policy makers need the criticality ranking of transportation network links so that they can act to suggest investment/improvement strategies. The criticality of transportation networks links is calculated through utilizing selected criticality metric(s) for various disruption scenarios. The disruption scenarios generally include link failure scenarios that can correspond to one link removal/degradation at a time (the most common) or simultaneous removal/degradation of multiple links. The common approach is to utilize singular component degradation/failure to calculate the individual criticality, i.e., run traffic assignment for each link removal scenario, calculate the difference of the network performance function (e.g., total system travel time) between the unaffected network and the failure scenario, and rank the links based on this functionality loss difference, i.e., the highest the difference, higher the ranking of the link. However, it is more likely that multiple links fail simultaneously, especially when considered within disaster conditions such as hurricanes and snowstorms. As shown in the literature, single link removals do not reveal the actual criticality of link(s) due to network dependencies, e.g., a link that may not create a large network performance change to be deemed critical, yet can cripple the system when fails in conjunction with others. In other words, the network dependencies make it difficult to isolate each link’s individual criticality. Multiple simultaneous link removal scenarios can capture the network interactions; however, the calculated criticality scores indicate the criticality ranking of scenarios than individual links, e.g., the links in scenario-1 is more critical than the links in scenario-2. In addition, running multiple link failure scenarios can be computationally infeasible due to the combinatorial nature of scenario creation. For example, for a medium size network with 100 links, the number of scenarios to include all simultaneous two-link removals are 100-2 = 4,950. For triple link removal, the number of scenarios increase to 100-3=485,100, and for quadruple removals, the total number of scenarios is 2,352,735. Considering that simultaneous quadruple link removals in a network size of 100 cannot reveal the network flow interdependencies, such multi-link failure analysis lead to impractical computation times. In this context, there are two main research questions: 1) How to calculate criticality of individual links based on scenarios that include multiple simultaneous failures?; 2) What is the optimal scenario generation approach that reveals the network flow dependencies while it is computationally tractable? For the first problem, PI Yazici has developed an approach that calculates individual link criticality based on a given number of scenarios. The approach utilizes the distribution of the criticality scores for each link, i.e., the criticality score distribution for link #X is composed of the criticality scores of scenarios that include link #X. The criticality ranking for each link is calculated based on its criticality score distribution’s mean, coefficient of variation and skewness. The approach was tested on real-life networks and it was shown that it provides robust link criticality rankings that account for network interactions. Hence, this project focuses on the optimal scenario generation strategy that will provide a systematic approach to select a smaller subset of all link failure scenarios that enables transportation criticality analysis that account for network flow interaction patterns with reasonable computation times. The developed procedure will also account for the network size and topology that affect network flow dependencies, thus the criticality of individual links.

Language

• English

Project

• Status: Active
• Funding: $89,618.00
• Contract Numbers:

  69A3552348321

• Sponsor Organizations:

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Managing Organizations:

  Florida A&M University, Tallahassee

  404 Foote/Hilyer
  Tallahassee, FL  United States  32307
• Project Managers:

  Moses, Ren

• Performing Organizations:

  Stony Brook University

  100 Nicolls Road
  Stony Brook, NY  United States  11794
• Principal Investigators:

  Yazici, M

• Start Date: 20240601
• Expected Completion Date: 20250531
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers Program

Subject/Index Terms

• TRT Terms: Network analysis (Planning); Service disruption; Traffic flow
• Subject Areas: Highways; Operations and Traffic Management; Planning and Forecasting;

Filing Info

• Accession Number: 01945641
• Record Type: Research project
• Source Agency: Rural Equitable and Accessible Transportation Center
• Contract Numbers: 69A3552348321
• Files: UTC, RIP
• Created Date: Feb 12 2025 5:56PM