This project aims to develop and test a novel human-cognizant decision-making framework for bridge maintenance that integrates predictive modeling and institutional trust, behavioral adaptation, and participatory insight to improve infrastructure resilience and expand long-term economic opportunity. Bridge maintenance decisions have wide-reaching societal consequences – affecting public safety, job accessibility, freight mobility, and economic productivity – yet current models often overlook the human and institutional dynamics that shape those outcomes. By embedding decision-maker trust in machine learning models for bridge deterioration prediction, modeling how individuals and businesses adapt to disruptions, and incorporating stakeholder input into maintenance prioritization, this research transforms how infrastructure decisions are made. The project supports the progress of science and engineering and serves the national interest by informing infrastructure strategies that enhance safety and deliver greater economic impact for the public. Technically, the project advances four key innovations: (1) it reframes trust as a measurable design element of machine learning models by testing how model explainability, data quality, and uncertainty influence adoption by public-sector agencies; (2) it introduces modeling of behavioral adaptation among commuters and businesses in response to bridge maintenance disruptions, drawing on theories of risk, habit, and loss aversion; (3)