Traffic Volume Benchmarks for Major Arterial Widening versus Expressway Construction: Exploratory Approach

Abstract

There exist several classes of physical interventions to address congestion at arterials, including widening (construction of additional lanes) and upgrading to expressway standards where possible. For each class of intervention, a key question is the threshold or benchmark traffic volume (BTV) at which the intervention is feasible. For long-term congestion mitigation planning for multiple arterial sections in a network and for planning-phase corridor development, agencies desire to identify BTVs for congestion mitigation interventions. The existing literature addresses this issue largely in feasibility analysis for specific projects, with little or no guidance regarding network-level planning considerations. In addressing this gap, this paper presents a detailed network-level planning methodology that is rooted in economic efficiency. The methodology first establishes three basic alternatives—Do-Nothing, Widening, and Upgrade-to-Expressway—and expresses the total lifecycle cost for each alternative as a function of the current traffic volume in terms of average daily traffic (ADT). The BTV is identified as the ADT at which the lifecycle costs of any two alternatives are equal. This paper demonstrates the application of the proposed methodology using data from in-service roads, and the results indicate that the relative weights assigned to the agency and user costs significantly influence the BTVs. In addition, results are presented for the scenario where these two cost categories are assigned equal weights; for this scenario, widening is never the best option. In the probabilistic situation, it is found that, at an 80% confidence level, it is economically attractive to upgrade a four-lane major arterial to an expressway when the ADT reaches 6,000; at the 100% confidence level, it is economically attractive to upgrade to an expressway when the ADT reaches 19,000.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 7 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.