01518782

Component

Links of a transportation network are subject to several forms of frequent non-severe disruptions like traffic incidents, snowing, flooding, road-space reallocations, road-space infiltrations, infrastructural failures, etc. Disregarding the cause, they may be viewed as multiple simultaneous disruptions on links’ capacities. Though the disruption or failure of a single link is widely studied, the area of multiple simultaneous disruptions on transportation networks is under-researched. Also, most of the existing studies assume probability distributions of capacity disruptions to be known. The objective of this paper is to study multiple simultaneous disruptions and to establish a consequent critical state. The authors assign different levels on links’ capacities and search for probability distributions over those levels that would result in a critical state. The critical state is modeled as the one which effects in poor network performance, and comparably equal expected path costs, that makes it hard for a user to identify the better path (i.e. it induces maximum indifference on expected utilities of alternative paths). The authors formulate the critical state link disruption problem as a minimax optimization problem, and solve it using a coevolutionary algorithm. The authors evaluate network performance in terms of expected system travel time. The formulation is exemplified by finding expected system travel time at the critical state on test networks. It is also shown how the performance at critical state can be used to measure network robustness.

This paper was sponsored by TRB committee ADB30(10) Paper Review Group #6.

01503729

14-4351

English

12p

2014

Transportation Research Board 93rd Annual Meeting

Digital/other

Figures; References; Tables

Algorithms; Distributions (Statistics); Network links; Networks; Optimization; Traffic incidents; Travel time

Road networks

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

Transportation Research Board Annual Meeting 2014 Paper #14-4351

TRIS, TRB, ATRI

Jan 27 2014 3:30PM