Road networks reliability estimations and optimizations: A Bi-directional bottom-up, top-down approach

Abstract

In this paper, we present a novel three-stage reduction-based estimation and optimization heuristics to evaluate and improve large-scale road networks’ reliability efficiently. Some necessary rules and tools are developed, in the 1st stage, to translate a large-scale physical road network into a mathematical graph. Then, in the 2nd stage, a bottom-up reduction algorithm is formulated to estimate network reliability. Finally, in the 3rd stage, a top-down method is proposed to improve the road network reliability by employing mathematical optimization models. Hence, this is a bidirectional bottom-up top-down (BUTD) method that (1) is genuinely based on the mathematical theory of reliability, (2) is computationally efficient by avoiding NP-hardness, and (3) by relaxing the need to employ pseudo metrics of reliability, e.g., vulnerability, resilience, etc. from the complex network theory in analyzing the large-scale road networks. The approach is explained using an illustrative example and a case study.