RSB/Collaborative Research: A Sequential Decision Framework to Support Trade Space Exploration of Multi-Hazard Resilient and Sustainable Building Designs

由于在决策过程中必须考虑到各种经济、环境、社会和技术因素，多灾害弹性和可持续建筑（RSB）的设计必然是复杂的。这些广泛的考虑导致了众多设计目标之间的权衡（例如，最小化生命周期成本、可能的损失和环境影响）。在这些权衡之间进行协商，需要新的决策框架和工具，以便对多危害RSB设计进行高效和有效的探索。行为经济学、心理学、市场营销和认知工程领域的文献都支持这样一个前提，即决策是构建和满足个人偏好的同时过程。本研究的目的是建立一个创新的顺序决策过程，为设计多危害RSB量身定制同时偏好构建和满意度。模型模拟将以增加保真度的顺序进行，同时决策者（dm）认识到设计空间和固有的权衡（交易空间），形成偏好，然后满足这些偏好，在整个过程中剔除考虑的设计集，以达到最终选择。该框架开发的关键如下：(1)创建一个复杂性指数，将建筑与其社会、技术、经济、政策和环境背景联系起来，并评估这些背景对建筑可持续性的影响，反过来，建筑对其更大的区域背景的影响；(2)迭代探索创新的多危害RSB设计；(3)建立可行的设计目标和设计指标的阈值。(4)开发贸易空间的数字可视化，以促进和证明设计选择。顺序决策框架将对综合设计过程产生积极的影响，它为dm、设计团队和利益相关者之间的学习、可视化和信息共享提供了机制，从而使结果是一个对环境、社会和财务负责的建筑设计。该研究项目的结果将是一个严格的、顺序的、基于渐进模型的仿真和可视化算法的决策框架，以支持多危害RSB设计的贸易空间探索。进化算法将在模型保真度的每个级别上使用，以生成考虑一系列材料，结构形式和建筑组件的土壤-基础-结构-围护结构建筑系统设计方案。通过整合现有的基于性能的概率评估模型、生命周期评估模型和新的概率建筑恢复模型，将为每个建筑设计方案和危险情景生成一系列广泛的设计度量。建筑恢复模型将使用系统可靠性方法开发，并将生成事件后功能和恢复时间指标。建筑恢复模型将明确地考虑内部功能和外部因素，例如公用事业和组织和技术系统的能力，从而允许在更广泛的社区弹性背景下理解特定的建筑设计。将开发新的交互式可视化分析技术和可视化算法，以促进多维贸易空间探索，使dm能够协商并深入了解冲突设计指标之间的复杂关系，以确定具有弹性和可持续性的设计。正式的框架将应用于对城市环境中受地震和/或飓风威胁的中高层住宅、商业和混合用途rsb的设计进行深入的案例研究。

The design of multi-hazard resilient and sustainable buildings (RSB) is necessarily complex due to the various economic, environmental, social, and technical considerations that must be factored into the decision-making process. These broad considerations result in tradeoffs among the numerous design objectives (e.g., minimizing life cycle cost, probable losses, and environmental impact). Negotiating these tradeoffs necessitates new decision-making frameworks and tools to enable efficient and effective exploration of multi-hazard RSB designs. Literature in the fields of behavioral economics, psychology, marketing, and cognitive engineering support the premise that decision-making is a simultaneous process of both constructing and satisfying one's preferences. The objective of this research is to formalize an innovative sequential decision process, tailored to simultaneous preference construction and satisfaction, for the design of multi-hazard RSB. Model simulations will be performed in a sequence of increasing fidelity while simultaneously decision-makers (DMs) become cognizant of the design space and inherent tradeoffs (trade space), form preferences, and then satisfy these preferences, culling the set of considered designs throughout to arrive at a final choice. Key to the framework's development are the following: (1) creating a complexity index that will situate the building in relation to its social, technical, economic, policy, and environmental context, and assessing the impact of this context on the sustainability of the building, and conversely, the impact of the building on its larger, regional context, (2) iteratively exploring for innovative multi-hazard RSB designs, (3) establishing sets of feasible design objectives and threshold values of design metrics, and (4) developing digital visualizations of the trade space to facilitate and justify design choices. The sequential decision framework will have a positive impact on the integrative design process by providing mechanisms for learning, visualizing, and sharing of information among the DMs, design team, and stakeholders so that the outcome is an environmentally, socially, financially responsible building design. The outcome of this research project will be a rigorous, sequential, decision framework based on progressive model-based simulation and visualization algorithms to support trade space exploration for multi-hazard RSB designs. Evolutionary algorithms will be employed at each level of model fidelity to generate soil-foundation-structural-envelope building system design alternatives, which consider an array of materials, structural forms, and building components. A broad array of design metrics will be generated for each building design alternative and hazard scenario(s) by integrating existing probabilistic performance-based assessment models, life cycle assessment models, and new probabilistic building recovery models. The building recovery models will be developed using a systems reliability approach and will generate the post-event functionality and recovery time metrics. The building recovery models will explicitly account for internal functions and externalities, such as utilities and the capacity of both organizational and technical systems, thereby allowing a particular building design to be understood in the broader context of a community's resilience. Novel interactive visual analytic techniques and visualization algorithms will be developed to facilitate multi-dimensional trade space exploration, allowing DMs to negotiate and gain insight into the intricate relationships among the conflicting design metrics to identify designs that are both resilient and sustainable. The formalized framework will be applied to in-depth case studies for the design of mid- and high-rise residential, commercial, and mixed-use RSBs threatened by earthquakes and/or hurricanes in urban environments.