26 November 2024 in IEEE Transactions on Intelligent Transportation Systems
Recent advancements in Large Language Models (LLMs) have ushered in opportunities to craft agents that exhibit human-like cognitive abilities, notably reasoning and planning. Leveraging the vast knowledge bases and sophisticated understanding of interrelations inherent in LLMs, this study explores their application in the dynamic and complex realm of traffic management. Specifically, we focus on Adaptive Traffic Control Systems (ATCS) as a critical use case, where designing effective traffic controllers parallels the cognitive tasks of reasoning and planning. This paper introduces a novel LLM-based agent development framework to create two types of traffic controllers: a Zero-Shot Chain of Thought approach and a Generally Capable Agent (GCA) approach. The latter integrates new knowledge from environmental interactions to enhance reasoning and planning capabilities. We implement and compare these controllers within a simulated traffic flow scenario at a single intersection using the Simulation of Urban Mobility (SUMO) against conventional traffic control methods, including fixed-time, gap-based, and delay-based controllers. Our results demonstrate that GCA-based controllers notably outperform traditional systems, reducing halted vehicle numbers by 48.03% and increasing average vehicle speed by 25.29%. These controllers also exhibit superior flexibility, generating diverse phase patterns that adapt dynamically to changing traffic conditions. This study underscores the transformative potential of LLMs in traffic management, promising significant enhancements in efficiency, responsiveness, and versatility in urban traffic control, ultimately improving urban life by mitigating traffic congestion’s economic and environmental impacts.
1 March 2023 in Journal of Management in Engineering
Decisions made at every stage of the built environment life cycle impact how materials in that environment are eventually managed at the end of their life, whether in a linear or cyclical manner. Linear fates often terminate in landfills, whereas more sustainable cyclical fates are continuously extended through reduction, reuse, repair, recycling, and recovery (5Rs) practices. Despite a notable increase in the literature on sustainable construction and demolition waste (CDW) management, there is still a lack of studies comprehensively addressing the full life cycle of the built environment. A gap also exists in the understanding of the root causes underlying unsustainable CDW management practices. To address these gaps and shortcomings, this study synthesizes the literature on CDW management to identify the causal factors contributing to the current unsustainable management practices, their root causes across the life cycle of the built environment, and sustainable strategies with proven efficacy in mitigating the detected root causes. Considering the implications of these strategies on the management of CDW, as well as their applicability across different stages of the built environment life cycle, this study characterizes them using the 5Rs waste hierarchy and using a five-stage built environment life cycle framework: interim, planning, construction, use, and end of life. All such information is integrated into a single ranking framework that proposes prioritized sustainable strategic pathways to address the most relevant root causes of unsustainable CDW management across the entire built environment life cycle. The actual implementation of the proposed framework for the sustainable management of the built environment is demonstrated through an analysis of Louisiana, in which peacetime construction/demolition and disaster debris generation and reconstruction/demolition have resulted in large amounts of CDW that often ended in landfills. Future decision makers can avoid similar fates by mitigating their own root causes of unsustainable waste management by better utilizing relevant mitigation strategies.
1 November 2026 in Journal of Management in Engineering
Infrastructure systems are influenced by a wide range of external factors, including demographic changes, economic cycles, environmental stressors, and technological transitions. These factors affect infrastructure demand, operating conditions, and deterioration processes through nonlinear, delayed, and interacting pathways, causing gradual shifts that can accumulate and emerge as abrupt changes in system performance at higher planning levels. This multiscale behavior creates significant challenges for infrastructure planning, as agencies need to monitor large volumes of heterogeneous data and derive timely, actionable insights from indicators reported at different temporal and spatial scales. Traditional planning methods, which often rely on isolated indicators or static forecasts, struggle to detect emerging patterns and diagnose their underlying drivers across system hierarchies. To address these limitations, this study proposes a hierarchy-aligned composite index framework that integrates diverse external factors and performance measures to support continuous monitoring and adaptive decision-making. The framework combines causal inference methods and dynamic factor analysis to identify influential drivers, extract latent common trends, and organize information in a structure consistent with infrastructure system governance. Rather than merely simplifying interpretation, the composite index streamlines abundant, multidimensional information into interpretable and traceable signals. A case study of Florida’s multimodal transportation system demonstrates the framework’s ability to track and diagnose system behavior across multiple transportation modes and planning levels. Practitioner review with FDOT planners confirms the interpretability and practical relevance of the indices and highlights the framework’s value for adaptive infrastructure planning under uncertainty.