A challenge systems engineers and designers face when applying system failure risk assessment methods such as Probabilistic Risk Assessment (PRA) during conceptual design is their reliance on historical data and behavioral models. This paper presents a framework for exploring a space of functional models using graph rewriting rules and a qualitative failure simulation framework that presents information in an intuitive manner for human-in-the-loop decision-making and human-guided design. An example is presented wherein a functional model of an electrical power system is iteratively perturbed to generate alternatives. The alternative functional models suggest different approaches to mitigating an emergent system failure vulnerability in the electrical power system’s the heat extraction capability. A preferred functional model configuration that has a desirable failure flow distribution can then be identified. The method presented here helps systems designers to better understand where failures propagate through systems and guides modification of systems functional models to adjust the way in which systems fail to have more desirable characteristics.

Detection of potential failures and human error and their propagation over time at an early design stage will help prevent system failures and adverse accidents. Hence, there is a need for a failure analysis technique that will assess potential functional/component failures, human errors, and how they propagate to affect the system overall. Prior work has introduced FFIP (Functional Failure Identification and Propagation), which considers both human error and mechanical failures and their propagation at a system level at early design stages. However, it fails to consider the specific human actions (expected or unexpected) that contributed towards the human error. In this paper, we propose a method to expand FFIP to include human action/error propagation during failure analysis so a designer can address the human errors using human factors engineering principals at early design stages. To explore the capabilities of the proposed method, it is applied to a hold-up tank example and the results are coupled with Digital Human Modeling to demonstrate how designers can use these tools to make better design decisions before any design commitments are made.

This paper presents a sequential Kriging optimization approach (SKO) for time-variant reliability-based design optimization (tRBDO) with the consideration of stochastic processes. To handle the extremely high dimensionality associated with time-variant uncertainties, stochastic processes are transformed to random parameters through the equivalent stochastic transformation, leading to equivalent time-independent reliability models that are capable of capturing system failures over time. To alleviate computational burden, Kriging-based surrogate modeling is utilized to predict the response of engineered systems. It is further integrated with Monte Carlo simulation (MCS) to approximate the probability of failure. To reduce the epistemic uncertainty due to the lack of data, a maximum confidence enhancement method (MCE) is employed to iteratively identify important points for updating surrogate models. Sensitivities of reliability with respect to design variables are estimated using the first-order score function in the proposed tRBDO framework. Two case studies are introduced to demonstrate the efficiency and accuracy of the proposed approach.

In reliability-based design optimization (RBDO), an optimal design which minimizes an objective function while satisfying a number of probabilistic constraints is found. As opposed to deterministic optimization, statistical uncertainties in design variables and design parameters have to be taken into account in the design process in order to achieve a reliable design. In the most widely used RBDO approaches, the First-Order Reliability Method (FORM) is used in the probability assessment. This involves locating the Most Probable Point (MPP) of failure, or the inverse MPP, either exactly or approximately. If exact methods are used, an optimization problem has to be solved, typically resulting in computationally expensive double loop or decoupled loop RBDO methods. On the other hand, locating the MPP approximately typically results in highly efficient single loop RBDO methods since the optimization problem is not necessary in the probability assessment. However, since all these methods are based on FORM, which in turn is based on a linearization of the deterministic constraints at the MPP, they may suffer inaccuracies associated with neglecting the nonlinearity of deterministic constraints. In a previous paper presented by the authors, the Response Surface Single Loop (RSSL) Reliability-based design optimization method was proposed. The RSSL-method takes into account the non-linearity of the deterministic constraints in the computation of the probability of failure and was therefore shown to have higher accuracy than existing RBDO methods. The RSSL-method was also shown to have high efficiency since it bypasses the concept of an MPP. In RSSL, the deterministic solution is first found by neglecting uncertainties in design variables and parameters. Thereafter quadratic response surface models are fitted to the deterministic constraints around the deterministic solution using a single set of design of experiments. The RBDO problem is thereafter solved in a single loop using a closed-form second order reliability method (SORM) which takes into account all elements of the Hessian of the quadratic constraints. In this paper, the RSSL method is used to solve the more challenging system RBDO problems where all constraints are replaced by one constraint on the system probability of failure. The probabilities of failure for the constraints are assumed independent of each other. In general, system reliability problems may be more challenging to solve since replacing all constraints by one constraint may strongly increase the non-linearity in the optimization problem. The extensively studied reliability-based design for vehicle crash-worthiness, where the provided deterministic constraints are general quadratic models describing the system in the whole region of interest, is used to demonstrate the capabilities of the RSSL method for problems with system reliability constraints.

The reliability of a system is usually measured by the probability that the system performs its intended function in a given period of time. Estimating such reliability is a challenging task when the probability of failure is rare and the responses are nonlinear and time variant. The evaluation of the system reliability defined in a period of time requires the extreme values of the responses in the predefined period of time during which the system is supposed to function. This work builds surrogate models for the extreme values of responses with the Kriging method. For the sake of computational efficiency, the method creates Kriging models with high accuracy only in the region that has high contributions to the system failure; training points of random variables and time are sampled simultaneously so that their interactions could be considered automatically. The example of a mechanism system shows the effectiveness of the proposed method.

Current methods of functional failure risk analysis do not facilitate explicit modeling of systems equipped with Prognostics and Health Management (PHM) hardware. As PHM systems continue to grow in application and popularity within major complex systems industries (e.g. aerospace, automotive, civilian nuclear power plants), implementation of PHM modeling within the functional failure modeling methodologies will become useful for the early phases of complex system design and for analysis of existing complex systems. Functional failure modeling methods have been developed in recent years to assess risk in the early phases of complex system design. However, the methods of functional modeling have yet to include an explicit method for analyzing the effects of PHM systems on system failure probabilities. It is common practice within the systems health monitoring industry to design the PHM subsystems during the later stages of system design — typically after most major system architecture decisions have been made. This practice lends itself to the omission of considering PHM effects on the system during the early stages of design. This paper proposes a new method for analyzing PHM subsystems’ contribution to risk reduction in the early stages of complex system design. The Prognostic Systems Variable Configuration Comparison (PSVCC) eight-step method developed here expands upon existing methods of functional failure modeling by explicitly representing PHM subsystems. A generic pressurized water nuclear reactor primary coolant loop system is presented as a case study to illustrate the proposed method. The success of the proposed method promises more accurate modeling of complex systems equipped with PHM subsystems in the early phases of design.

Optimizing the topology of complex infrastructure systems can minimize the impact of cascading failures due to an initiating failure event. This paper presents a novel approach for the concept-stage design of complex infrastructure systems by integrating model-based design with network analysis to increase system robustness. This approach focuses on system performance after cascading has occurred, and examines design trade-offs of the resultant (or degraded) system state. In this research, robustness is defined as the invariability of system performance due to uncertain failure events. Where a robust network has the ability to meet minimum performance requirements despite the impact of cascading failures. This research is motivated by catastrophic complex infrastructure system failures such as the August 13th Blackout of 2003, highlighting the vulnerability of systems such as the North American Power Grid (NAPG). A mathematical model was developed using an adjacency matrix, where removing a network connection simulates uncertain failure events. Performance degradation is iteratively calculated as failures cascade throughout the system, and robustness is measured by the lack of performance variability over multiple cascading failure scenarios. Two case studies are provided: an extrapolated IEEE 14 test bus, and the Oregon State University campus power network. The overarching goal of this research is to understand key system design trade-offs between robustness, performance objectives, and cost. In addition, optimizing network topologies to mitigate performance loss during concept-stage design will enable system robustness.

Evidence gathered from industry indicates that railway coupling system failures have become a limitation for further developments of heavy haul trains. Friction draft gears have implications for both longitudinal train dynamics and rolling stock fatigue; therefore, optimization of friction draft gears could be a possible solution to conquer the limitation. In this paper, a methodology for optimization of friction draft gear design based on an advanced friction draft gear model is proposed. The methodology proposes using simulation techniques such as longitudinal train dynamics simulation and a Genetic Algorithm to develop improved parameters.

This paper presents a novel safety specification and verification approach based on the compositional reasoning and model checking algorithms. The behavioral specification of each component and subsystem is modeled to describe the overall structure of the design. Then, these specifications are analyzed to determine the least number of component redundancies that are required to tolerate and prevent catastrophic system failure. The framework utilizes Labelled Transition Systems (LTS) formalism to model the behavior of components and subsystems. Furthermore, compositional analysis is used to reason about the components’ constraints (or assumptions) on their environments and the properties (or guarantees) of their output. This identification of local safety properties of components and subsystems leads to satisfaction of the desired safety requirements for the global system. A model of quad-redundant Electro-Mechanical Actuator (EMA) is constructed and, in an iterative approach, its safety properties are analyzed. Experimental results confirm the feasibility of the proposed approach for verifying the safety issues associated with complex systems in the early stages of the design process.

Lifecycle health management plays an increasingly important role in realizing resilience of aging complex engineered systems since it detects, diagnoses, and predicts system-wide effects of adverse events, therefore enables a proactive approach to deal with system failures. To address an increasing demand to develop high-reliability low-cost systems, this paper presents a new platform for operational stage system health management, referred to as Evolving Design Model Synchronization (EDMS), which enables health management of aging engineered systems by efficiently synchronizing system design models with degrading health conditions of actual physical system in operation. A Laplace approximation approach is employed for the design model updating, which can incorporate heterogeneous operating stage information from multiple sources to update the system design model based on the information theory, thereby increases the updating accuracy compared with traditionally used Bayesian updating methodology. The design models synchronized over time using sensory data acquired from the system in operation can thus reflect system health degradation with evolvingly updated design model parameters, which enables the application of failure prognosis for system health management. One case study is used to demonstrate the efficacy of the proposed approach for system health management.

This paper presents an integrated performance measure approach (iPMA) for system reliability assessment considering multiple dependent failure modes. An integrated performance function is developed to envelope all component level failure events, thereby enables system reliability approximation by considering only one integrated system limit state. The developed integrated performance function possesses two critical properties. First, it represents exact joint failure surface defined by multiple component failure events, thus no error will be induced due to the integrated limit state function in system reliability computation. Second, smoothness of the integrated performance on system failure surface can be guaranteed, therefore advanced response surface techniques can be conveniently employed for response approximation. With the developed integrated performance function, the maximum confidence enhancement based sequential sampling method is adopted as an efficient component reliability analysis tool for system reliability approximation. To furthermore improve the computational efficiency, a new constraint filtering technique is developed to adaptively identify active limit states during the iterative sampling process without inducing any extra computational cost. One case study is used to demonstrate the effectiveness of system reliability assessment using the developed iPMA methodology.

This paper presents a new system design platform and approaches leading to the development of resilient engineered systems through integrating design of system functions and prognosis of function failures in a unified design framework. Failure prognosis plays an increasingly important role in complex engineered systems since it detects, diagnoses, and predicts the system-wide effects of adverse events, therefore enables a proactive approach to deal with system failures at the life cycle use phase. However, prognosis of system functional failures has been largely neglected in the past at early system design stage, mainly because quantitative analysis of failure prognosis in the early system design stage is far more challenging than these activities themselves that have been mainly carried out at the use phase of a system life cycle. In this paper, a generic mathematical formula of resilience and predictive resilience analysis will be introduced, which offers a unique way to consider lifecycle use phase failure prognosis in the early system design stage and to systematically analyze their costs and benefits, so that it can be integrated with system function designs concurrently to generate better overall system designs. Engineering design case studies will be used to demonstrate the proposed design for resilience methodology.

Many embedded systems applications have hard timing requirements where real-time processes with precedence and exclusion relations must be completed before specified deadlines. This requires that the worst-case computation times of the real-time processes be estimated with sufficient precision during system design, which sometimes can be difficult in practice. If the actual computation time of a real-time process during run-time exceeds the estimated worst-case computation time, an overrun will occur, which may cause the real-time process to not only miss its own deadline, but also cause a cascade of other real-time processes to also miss their deadline, possibly resulting in total system failure. However, if the actual computation time of a real-time process during run-time is less than the estimated worst-case computation time, an underrun will occur, which may result in under-utilization of system resources. This paper describes a method for handling underruns and overruns when scheduling a set of real-time processes with precedence and exclusion relations using a pre-run-time schedule. The technique effectively tracks and utilizes unused processor time resources to reduce the chances of missing real-time process deadlines, thereby providing the capability to significantly increase both system utilization and system robustness in the presence of inaccurate estimates of the worst-case computation times of real-time processes.

A recently developed technique of identifying system topology using tests and system simulation for reliability applications has remanufacturing implications. Two issues affect successful remanufacturing of products; first, lack of knowledge about what constitutes failure of a product in the user’s mind, and second, inability to determine if system failure has occurred based on failure of some components even if all failure modes are known. Remanufacturing must account for soft failures , failures that do not affect the functionality of a product but the overall appeal to the customer. In this paper, we propose defining multiple perception-based product attributes and treating them as fictitious components of a reliability block diagram. Using a recently developed method, we then infer the system topology in a reliability engineering context in terms of success or failure of these components. This allows us to identify the actual or impending failure of the product which can be critical in determining whether the product should be recovered or sold to another customer with different requirements. We demonstrate our proposed method on a cell phone reuse problem.

Advances in high performance sensing and signal processing technology enable the development of failure prognosis tools for wind turbines to detect, diagnose, and predict the system-wide effects of failure events. Although prognostics can provide valuable information for proactive actions in preventing system failures, the benefits have not been fully utilized for the operation and maintenance decision making of wind turbines. This paper presents a generic failure prognosis informed decision making tool for wind farm operation and maintenance while considering the predictive failure information of individual turbine and its uncertainty. In the presented approach, the probabilistic damage growth model is used to characterize individual wind turbine performance degradation and failure prognostics, whereas the economic loss measured by monetary values and environmental performance measured by unified carbon credits are considered in the decision making process. Based on the customized wind farm information inputs, the developed decision making methodology can be used to identify optimum and robust strategies for wind farm operation and maintenance in order to maximize the economic and environmental benefits concurrently. The efficacy of proposed prognosis informed maintenance strategy is compared with the condition based maintenance strategy and demonstrated with the case study.

System health diagnostics provides diversified benefits such as improved safety, improved reliability and reduced costs for the operation and maintenance of engineered systems. Successful health diagnostics requires the knowledge of system failures. However, with an increasing complexity it is extraordinarily difficult to have a well-tested system so that all potential faulty states can be realized and studied at product testing stage. Thus, real time health diagnostics requires automatic detection of unexampled faulty states through the sensory signals to avoid sudden catastrophic system failures. This paper presents a hybrid inference approach (HIA) for structural health diagnosis with unexampled faulty states, which employs a two-fold inference process comprising of preliminary statistical learning based anomaly detection and artificial intelligence based health state classification for real time condition monitoring. The HIA is able to identify and isolate the unexampled faulty states through interactively detecting the deviation of sensory data from the known health states and forming new health states autonomously. The proposed approach takes the advantages of both statistical approaches and artificial intelligence based techniques and integrates them together in a unified diagnosis framework. The performance of proposed HIA is demonstrated with a power transformer and roller bearing health diagnosis case studies, where Mahalanobis distance serves as a representative statistical inference approach.

For complex, safety-critical systems failures due to component faults and system interactions can be catastrophic. One aspect of ensuring a safe system design is the analysis of the impact and risk of potential faults early in the system design process. This early design-stage analysis can be accomplished through function-based reasoning on a qualitative behavior simulation of the system. Reasoning on the functional effect of failures provides designers with the information needed to understand the potential impact of faults. This paper proposes three different methods for evaluating and grouping the results of a function failure analysis and their use in design decision-making. Specifically, a method of clustering failure analysis results based on consequence is presented to identify groups of critical failures. A method of clustering using Latent Class Analysis provides characterization of high-level, emergent system failure behavior. Finally, a method of identifying functional similarity provides lists of similar and identical functional effects to a system state of interest. These three methods are applied to the function-based failure analysis results of 677 single and multiple fault scenarios in an electrical power system. The risk-based clustering found three distinct levels of scenario functional impact. The Latent Class Analysis identified five separate failure modes of the system. Finally, the similarity grouping identified different groups of scenarios with identical and similar functional impact to specific scenarios of interest. The overall goal of this work is to provide a framework for making design decisions that decrease system risks.

Traditional data-driven prognostics often requires a large amount of failure data for the offline training in order to achieve good accuracy for the online prediction. However, in many engineered systems, failure data are fairly expensive and time-consuming to obtain while suspension data are readily available. In such cases, it becomes essentially critical to utilize suspension data, which may carry rich information regarding the degradation trend and help achieve more accurate remaining useful life (RUL) prediction. To this end, this paper proposes a co-training-based data-driven prognostic algorithm, denoted by Coprog , which uses two individual data-driven algorithms with each predicting RULs of suspension units for the other. The confidence of an individual data-driven algorithm in predicting the RUL of a suspension unit is quantified by the extent to which the inclusion of that unit in the training data set reduces the sum square error (SSE) in RUL prediction on the failure units. After a suspension unit is chosen and its RUL is predicted by an individual algorithm, it becomes a virtual failure unit that is added to the training data set. Results obtained from two case studies suggest that Coprog gives more accurate RUL predictions compared to any individual algorithm without the consideration of suspension data and that Coprog can effectively exploit suspension data to improve the accuracy in data-driven prognostics.

Drivetrains are a major source of vibration, noise and system failures. Accordingly, a significant amount of time and effort is being invested developing simulation methods in order to better understand and avoid potentially damaging vibrations, even before prototypes are created for testing. The first step in simulating any drivetrain is creating suitable virtual models to investigate particular phenomena. Too much model detail leads to long computation times and difficulties in interpreting results, while too little may fail to include desired effects. Because the various levels of detail available in multi-body simulation (MBS) are practically limitless, a significant amount of attention must be given in order to choose the appropriate modeling elements. In the simplest form an entire drivetrain can be modeled as several rigid masses connected with torsional springs, which is justifiable for fundamental concept analyses. For other analyses, full three dimensional modeling with complex components may be necessary. Higher frequency analyses may even necessitate the inclusion of material bending for achieving accurate results. The various available elements for modeling specific components must be well understood in order that appropriate choices are made. Modeling requirements for the elements commonly used in the simulation of drivetrains will be discussed. For example: bearings, gearwheels, universal and constant velocity joints, frequency and amplitude dependent mounts, flexible components (e.g. shafts and gearbox housings), etc. Once virtual models are available, various analysis methods are applied in order to aid designers in identifying and quantifying potentially damaging vibrations. Again the application and limitation of these methods must be well understood in order to generate meaningful results. The following methods will be compared and discussed: resonance analysis, linear system analysis, run-up Fast Fourier Transformation analysis, order analysis, transfer path analysis and durability analysis. These drivetrain modeling techniques and analysis methods are not limited to any specific field of engineering, but can be applied to an extensive range of engineering disciplines. Analyses applied to virtual models out of the automotive and wind turbine sectors will be shown.

A reliability analysis method is presented for time-dependent systems under uncertainty. The system response is modeled as a parameterized random process. A double-loop optimization algorithm is used. The inner loop calculates the maximum response in time using a global-local search method, and transforms a time-dependent problem into a time-independent one. The outer loop calculates multiple most probable points (MPPs) which are commonly encountered in vibration problems. The dominant MPPs with the highest contribution to the probability of failure are identified. A niching genetic algorithm is used because of its ability to simultaneously identify multiple solutions. All potential MPPs are initially identified approximately and their location is efficiently refined using a gradient-based optimizer with local metamodels. Among all MPPs, the significant ones are identified using a correlation analysis. Approximate limit states are built at the identified MPPs, and the system failure probability is estimated using bi-modal bounds. The vibration response of a cantilever plate under random oscillating pressure load and a point load illustrates the proposed method. The finite-element model is used to calculate the response.