A. James Clark School of Engineering
Permanent URI for this communityhttp://hdl.handle.net/1903/1654
The collections in this community comprise faculty research works, as well as graduate theses and dissertations.
Browse
2 results
Search Results
Item HYBRID RESILIENCE FRAMEWORK FOR SYSTEMS OF SYSTEMS INCORPORATING STAKEHOLDER PREFERENCES(2018) Emanuel, Roy Nelson; Ayyub, Bilal; Reliability Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)From Presidential Policy Directive 21, to professional societies’ national meetings, to major United Nations initiatives, stakeholders recognize the value of achieving resilient systems. The literature clamors with methods to assess resilience of systems quantitatively and qualitatively. Resilience models typically focus on system performance and the threat to the system. Few models consider the preferences of the stakeholders of the systems. This course of study identified three gaps in the literature: first, the focus on system performance without considering the preferences of stakeholders; second, lack of resilience model-to-model comparison; and third, lack of a common framework for applying resilience models across domains and systems of systems. This course of study investigated the impact of incorporating stakeholder preferences into four existing resilience models: Resilience Factor, Quotient Resilience, Total Quotient Resilience, and Integral Resilience. The incorporated stakeholder preferences were time horizon, endogenous performance preference, and intertemporal substitutability of system performance. An analysis of the resultant eight illustrative models showed the models' comparative sensitivity to changes in system performance and stakeholder preferences using four fundamental system performance and stakeholder preference models. A deterministic system dynamics model of a city's critical infrastructure provided inputs to the eight models for an initial case study. The first phase identifies three stakeholder preference profiles for the water delivery infrastructure. The second phase assesses the impact of electrical outages on seven other critical infrastructures. The results of the sensitivity analysis and the initial case study led to selection of the Extended Integral Resilience model for additional demonstrations. Stochastic inputs for the system dynamics model showed a range of resilience outcomes for each stakeholders' infrastructure for five courses of action. The hybrid resilience model used Department of Energy reports on Puerto Rico's recovery from Hurricane Maria to generate a resilience value. A discrete event simulation of a fleet of aircraft used to train aviators provided the basis for the second set of case studies. The study considered the points of view of the Squadron Commanders which were limited to three year increments, and the program manager which considered a thirty-five year time horizon. The functional outputs of the model were graduates per quarter, aircraft ready to fly each day, and satisfied graduates per quarter. The case study introduced and demonstrated an event and time dependent intertemporal substitutability algorithm to be defined by the stakeholder.Item ADVANCES IN SYSTEM RELIABILITY-BASED DESIGN AND PROGNOSTICS AND HEALTH MANAGEMENT (PHM) FOR SYSTEM RESILIENCE ANALYSIS AND DESIGN(2011) Hu, Chao; Youn, Byeng D.; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Failures of engineered systems can lead to significant economic and societal losses. Despite tremendous efforts (e.g., $200 billion annually) denoted to reliability and maintenance, unexpected catastrophic failures still occurs. To minimize the losses, reliability of engineered systems must be ensured throughout their life-cycle amidst uncertain operational condition and manufacturing variability. In most engineered systems, the required system reliability level under adverse events is achieved by adding system redundancies and/or conducting system reliability-based design optimization (RBDO). However, a high level of system redundancy increases a system's life-cycle cost (LCC) and system RBDO cannot ensure the system reliability when unexpected loading/environmental conditions are applied and unexpected system failures are developed. In contrast, a new design paradigm, referred to as resilience-driven system design, can ensure highly reliable system designs under any loading/environmental conditions and system failures while considerably reducing systems' LCC. In order to facilitate the development of formal methodologies for this design paradigm, this research aims at advancing two essential and co-related research areas: Research Thrust 1 - system RBDO and Research Thrust 2 - system prognostics and health management (PHM). In Research Thrust 1, reliability analyses under uncertainty will be carried out in both component and system levels against critical failure mechanisms. In Research Thrust 2, highly accurate and robust PHM systems will be designed for engineered systems with a single or multiple time-scale(s). To demonstrate the effectiveness of the proposed system RBDO and PHM techniques, multiple engineering case studies will be presented and discussed. Following the development of Research Thrusts 1 and 2, Research Thrust 3 - resilience-driven system design will establish a theoretical basis and design framework of engineering resilience in a mathematical and statistical context, where engineering resilience will be formulated in terms of system reliability and restoration and the proposed design framework will be demonstrated with a simplified aircraft control actuator design problem.