A Strategic Move for Long-Term Bridge Performance within a Game Theory Framework by a Data-Driven Co-Active Mechanism

Abstract

The Federal Highway Administration (FHWA) requires that states have less than 10% of the total deck area that is structurally deficient. It is a minimum risk benchmark for sustaining the National Highway System bridges. Yet, a decision-making framework is needed for obtaining the highest possible long-term return from investments on bridge maintenance, rehabilitation, and replacement (MRR). This study employs a data-driven coactive mechanism within a proposed game theory framework, which accounts for a strategic interaction between two players, the FHWA and a state Department of Transportation (DOT). The payoffs for the two players are quantified in terms of a change in service life. The proposed framework is used to investigate the element-level bridge inspection data from four US states (Georgia, Virginia, Pennsylvania, and New York). By reallocating 0.5% (from 10% to 10.5%) of the deck resources to expansion joints and joint seals, both federal and state transportation agencies (e.g., FHWA and state DOTs in the U.S.) will be able to improve the overall bridge performance. This strategic move in turn improves the deck condition by means of a co-active mechanism and yields a higher payoff for both players. It is concluded that the proposed game theory framework with a strategic move, which leverages element interactions for MRR, is most effective in New York where the average bridge service life is extended by 15 years. It is also concluded that the strategic move can lead to vastly different outcomes. Pennsylvania’s concrete bridge management strategy currently appears to leverage a co-active mechanism in its bridge MRR strategies. This is noteworthy because its bridges are exposed to similar environmental conditions to what is obtainable in Virginia and New York and are subjected to more aggressive weather conditions than those in Georgia. This study illustrates how a strategic move affects the payoffs of different players by numerically quantifying changes in service life from bridge time-dependent bridge performance relationships.