Optimization of Degradable Road Network Considering VMS Information and Heterogeneous ATIS Users

Abstract

Optimization of a degradable road network with two types of information facilities (on-road and onboard) is investigated under recurrent congestion. The key elements of this paper are the adjustment of network structure (i.e., degradable links) and a four-level game about the relationships among the optimization objective, the penetrability of onboard advanced traveler information systems (ATISs), the information utility of variable message signs (VMSs), and hybrid traffic assignment. The objectives of this study are twofold: (1) to study the impact of road network structure, ATIS, and VMS on transport performance, and (2) to improve transport performance by modifying road network structure and ATIS market share. First, it is assumed that hybrid traffic can employ real-time ATIS and VMS information. If VMS facilities are installed and function on some main roads, a VMS attenuation factor and traffic information quantity are formulated and employed to define the coefficient for VMS impact on vehicles. Using the impact coefficient and link disutility, a new formulation of link “length” for path-size attributes is proposed. Second, based on a path-size logit (PSL)–stochastic user equilibrium model, multilevel programming to maximize travel time reliability via road link capacity degradation (two types of degradation are differentiated: active and passive) and ATIS market share control is established. Third, a path-based solution algorithm is established by embedding the method of successive averages within the framework of an evolutionary algorithm. The model formulation and algorithm are analyzed using the Nguyen and Dupuis network. Within the framework of the evolutionary algorithm, investigations of several scenarios indicate that transport performance can be significantly enhanced by setting the remaining road links to be potentially degradable after the inevitably degraded links are identified, adding VMSs to the transport network, and exogenously determining the penetration and compliance rates of ATISs. For the three matrices of relatively high general demand, the travel time reliability of the specific origin-destination (OD) pair can be improved by rates of 29.83, 31.36, and 56.54%, and the total expected travel time of the whole network and the specific OD pair is reduced.

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 — 2 citations exist upstream.

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