CAREER: An Integrated Framework for Infrastructure Management: Exploiting Advanced Inspection Technologies to Support Condition Assessment, Forecasting and Decision-Making

美国西北大学Pablo L. Durango-Cohen教授，职业生涯：基础设施管理的集成框架：利用先进的检测技术来支持状态评估、预测和决策摘要：该提案涉及开发一个集成分析框架来支持基础设施系统的管理。该研究方法认识到管理过程的复杂性，并提供了一个全面的系统级框架，同时处理数据处理、性能和优化建模。该研究将一类复杂的优化工具（随机最优控制理论）与严格的统计方法（时间序列模型的状态空间规范）联系起来。这种整合允许提议的框架作为一种实用的工具，利用先进检查技术的广泛能力来支持管理需求，以及帮助选择和部署检查技术。该研究包括时间序列模型的状态空间规范的制定和估计，以支持状态评估和预测，以及优化模型的制定和分析，以支持维修和修复决策。时间序列模型的状态空间规范的发展对研究基础设施的劣化和检查具有重要的贡献。这些模型提供了一种创新的方法来研究与位置、环境、设计和建造、交通量和混合以及维护活动有关的因素对基础设施系统性能的影响。与优化组件的无缝链接意味着该框架可用于根据经济标准评估和优化决定这些外生因素的决策。制定有效和高效率的政策，分配资源以保存和改善民用基础设施，例如路面网，对国家的持续社会和经济繁荣和增长至关重要。这是由于基础设施恶化/不足所造成的深远和严重的负面影响，以及每年数百亿美元的管理开支规模。提出这项研究的动机是，传感和通信/信息技术的进步导致了复杂仪器的扩散，这些仪器用于评估和测量基础设施状况及其导致其退化的因素，并且需要解决基本的计算和统计挑战，以便利用新获得的大量和广泛的信息。拟议的研究是长期大规模努力的一个方面，旨在开发可应用于电信和物流等不同领域的网络的建模、分析和控制方法。此外，该提案还描述了如何将研究工作整合到教育计划中，旨在通过开发和传播教育材料、推广和学生研究参与，以及努力建立工程领域的多样性，弥合学术研究与最先进实践之间的差距。拟议计划的一个值得注意的特点是建立在与运输机构和主要服务提供者的既定合作基础上，以确保研究和教育活动解决重大问题，并确保结果是实际的和可执行的。

CMS 0547471CAREER: An Integrated Framework for Infrastructure Management: Exploiting Advanced Inspection Technologies to Support Condition Assessment, Forecasting and Decision-Making Pablo L. Durango-Cohen, Northwestern UniversityAbstract:This proposal addresses the development of an integrated analytical framework to support the management of infrastructure systems. The research approach recognizes the complex nature of the management process, and provides a comprehensive, system-level framework to simultaneously address data processing, performance and optimization modeling. The research links a general class of sophisticated optimization tools (stochastic optimal control theory) with rigorous statistical methods (state-space specifications of time series models). This integration allows the proposed framework to serve as a practical tool exploit the extensive capabilities of advanced inspection technologies to support management needs, as well as to aid in the selection and deployment of inspection technologies. The research consists of the formulation and estimation of state-space specifications of time series models to support condition assessment and forecasting, and the formulation and analysis of optimization models to support maintenance and rehabilitation decision-making. The development of state-space specifications of time series models represents a significant contribution to study infrastructure deterioration and inspection. These models provide an innovative approach to study the effect of factors related to location, environment, design and construction, traffic volume and mix, and maintenance activities on the performance of infrastructure systems. The seamless link with the optimization component means that the framework can then be used to evaluate and optimize decisions that determine these exogenous factors based on economic criteria. The development of effective and efficient policies to allocate resources for the preservation and improvement of civil infrastructure, e.g., pavement networks, is crucial to the nations continued social and economic prosperity and growth. This is due to both the far-reaching and serious negative impacts of deteriorating/deficient infrastructure, as well as the scale of management expenditures, which are on the order of tens of billions of dollars per year. The proposed research is motivated by advances in sensing and communications/information technologies that have resulted in the proliferation of sophisticated instruments to evaluate and measure infrastructure condition and the factors that cause its degradation, and by the need to address fundamental computational and statistical challenges in order exploit the abundance and breadth of newly available information. The proposed research is one facet of a long-term, large-scale effort to develop modeling, analysis and methods of control that can be applied to networks in fields as diverse as telecommunications and logistics. In addition, the proposal describes how the research efforts will be integrated into an educational plan aimed at bridging the gap that separates academic research from state-of-the-art practice through the development and dissemination of educational materials, outreach and student research involvement, and efforts to build diversity in engineering. A noteworthy feature of the proposed plan is to build on established collaborations with transportation agencies and leading service providers in order to ensure that the research and educational activities address significant problems, as well as to ensure that the results are practical and implementable.