Classes Of Failures Research Articles

Use of Human-Automation Taxonomies for System Modeling

Skraaning and Jamieson indicate the importance of developing a definition of automation failure in order to guide development of human performance models to predict human-automation interactions. They go on to develop a taxonomy including four categories: Elementary automation failures; Systemic automation failures; Human-automation breakdowns; and Human and organizational slips and misconceptions. They suggest that the latter category be excluded from classes of automation failures. They then proceed to identify important factors within each category. This paper presents a complementary taxonomy designed to identify factors contributing to automation-related failures in order to develop probes to elicit predictions of failure stories for some proposed new system design from a range of experts. While there is overlap between these two taxonomies, there are important differences. It is suggested that they could be integrated or used jointly both for knowledge elicitation and to identify requirements regarding needed capabilities of predictive performance models.

Methodological foundation for improving the quality of intelligent decision-making system software

On the basis of a detailed analysis, existing terminological interpretations of the concept of "software quality" have been generalized, conclusions are drawn about the correspondence of the terms used to assess the quality of general software in the process of assessing the quality of software of intelligent decision-making systems (IDMS). It has been proved that the quality of the IDMS software is a complex multi-criteria indicator that takes into account not only the performance of the individual software module as a subsystem, but also the causal relationships of the elements of the software system itself. The main differences in software quality assessment between the functional and formal approaches are shown. The structure of the criterion of guarantor capacity of decision-making systems software has been investigated and conclusions have been drawn on the influence of its main components on the evaluation of IDMS software and on the provision of reliable computing process. On the basis of the analysis of the list of attributes and the quality metric of the IDMS software, it is established that the guarantee is determined by the reliability of the software structure itself and is characterised by the restoration of the functional state after failures or failures. The interrelationship and influence of IDMS software design quality indicators on the characteristics and sub-characteristics of the IDMS software is established, an example of the interrelationship between characteristics (factors) and quality indicators, the method of measuring quality indicators and design processes is given. On the basis of the conducted research, IDMS software denial regimes have been defined and their impact on the decision-making process has been shown. Detailed classes of failures and their influence on compliance of IDMS software with the task of development are shown. It has been shown that the reliability of IDMS is a dynamic concept, manifested in time, and is strongly dependent on the presence / absence of defects in the interaction. A detailed analysis of methods of software quality assurance and control has been carried out, and conclusions have been drawn on the possibility of their application IDMS software. The maturity model of the IDMS software has been improved and validated, and the maturity structure of the software as an indicator of the quality of the IDMS has been introduced.

Enhancing Robustness of On-line Learning Models on Highly Noisy Data

Classification algorithms have been widely adopted to detect anomalies for various systems, e.g., IoT, cloud and face recognition, under the common assumption that the data source is clean, i.e., features and labels are correctly set. However, data collected from the wild can be unreliable due to careless annotations or malicious data transformation for incorrect anomaly detection. In this paper, we extend a two-layer on-line data selection framework: Robust Anomaly Detector (RAD) with a newly designed ensemble prediction where both layers contribute to the final anomaly detection decision. To adapt to the on-line nature of anomaly detection, we consider additional features of conflicting opinions of classifiers, repetitive cleaning, and oracle knowledge. We on-line learn from incoming data streams and continuously cleanse the data, so as to adapt to the increasing learning capacity from the larger accumulated data set. Moreover, we explore the concept of oracle learning that provides additional information of true labels for difficult data points. We specifically focus on three use cases, (i) detecting 10 classes of IoT attacks, (ii) predicting 4 classes of task failures of big data jobs, and (iii) recognising 100 celebrities faces. Our evaluation results show that RAD can robustly improve the accuracy of anomaly detection, to reach up to 98.95% for IoT device attacks (i.e., +7%), up to 85.03% for cloud task failures (i.e., +14%) under 40% label noise, and for its extension, it can reach up to 77.51% for face recognition (i.e., +39%) under 30% label noise. The proposed RAD and its extensions are general and can be applied to different anomaly detection algorithms.

Effectiveness and safety of secukinumab for psoriasis in real-world practice: analysis of subgroups stratified by prior biologic failure or reimbursement.

Background:Little is known about the treatment outcomes of secukinumab in clinical practice, which differ from those in clinical trials. The effectiveness of biologics may differ in psoriasis patients with previous biologics exposure. The objective of this study was to investigate the real-world effectiveness and safety of secukinumab therapy and analyze subgroups stratified by reimbursement or prior biologic failure.Methods:This retrospective multicenter study collected data from a cohort of 118 consecutive patients who received secukinumab treatment between December 2015 and March 2018. Effectiveness was evaluated by degree of improvement in the Psoriasis Area and Severity Index (PASI) scores. Adverse events and reasons for discontinuation were also recorded.Results:The mean PASI improvement rate at weeks 4, 12, 24, and 36 was 63.5%, 77.7%, 78.7%, and 76.0%, respectively. Compared with reimbursed patients, nonreimbursed patients had a significantly lower baseline PASI and a shorter mean disease duration of psoriasis; they were more frequently biologic-naïve, had used less prior traditional antipsoriatic drugs and were more likely to be treated with secukinumab 150 mg. The effectiveness of secukinumab in nonreimbursed patients was superior despite higher discontinuation rates. Compared with patients without prior biologic failure, patients with prior biologic failure had a significantly lower mean PASI improvement at weeks 12, 24, 36, and 48. The decline in response rates to secukinumab tended to be more pronounced for patients who failed ustekinumab than tumor necrosis factor-α inhibitors. Moreover, the number of prior biologic failures was associated with a decreased response rate and increased likelihood of secondary loss of effectiveness of secukinumab therapy.Conclusion:In a real-life clinical setting, the characteristics of nonreimbursed patients receiving secukinumab treatment differed from those of reimbursed patients. The PASI improvement for secukinumab was substantial but lower than that in clinical trials. The number and classes of prior biologic failures impact the treatment response to secukinumab.

A Diagnosis Study on a Train Passenger Access System using Petri Net Models

In this paper, we conduct a diagnosis analysis of the passenger access system, while considering a high level abstraction perspective that allows for adapting discrete event models to represent the system behavior. Firstly, we establish Petri net behavioral models for the global system functions, including the nominal operating mode and various faulty behaviors. Then, based on the established Petri net models, a diagnoser-based approach is brought into play to investigate the diagnosability of the system regarding the different predetermined classes of failures.

Model-Based Monitoring of a Train Passenger Access System

The passenger access system (PAS) is a complex mechatronic train onboard module with high reliability and safety requirements. This module fulfills one of the dozen main onboard functions onboard train. Consequently, any related fault occurrence may have a serious impact on the safety and availability of the whole train operation. In this context, developing effective automated monitoring and diagnostic techniques for the PAS, as early as from the design phase of the system, becomes an essential and challenging task. In this paper, we carry out a monitoring study on this system, while considering a sufficiently high-level abstraction perspective that allows for adapting discrete event models representing the behavior of the system. First, we establish a Petri net behavioral model that includes the nominal operating mode as well as various faulty behaviors. Then, based on the established Petri net models, a fault detection approach is used to investigate the diagnosability property and synthesize the diagnosers regarding different predetermined classes of failures. Finally, we show how the outputs of the diagnosability analysis can help make efficient design choices that allow for improving the safety of the whole system.

Why does the orientation change mess up my Android application? From GUI failures to code faults

SummaryThis paper investigates the failures exposed in mobile apps by the mobile‐specific event of changing the screen orientation. We focus on GUI failures resulting in unexpected GUI states that should be avoided to improve the apps quality and to ensure better user experience. We propose a classification framework that distinguishes 3 main classes of GUI failures due to orientation changes and exploit it in 2 studies that investigate the impact of such failures in Android apps. The studies involved both open‐source and apps from Google Play that were specifically tested exposing them to orientation change events. The results showed that more than 88% of these apps were affected by GUI failures, some classes of GUI failures were more common than others, and some GUI objects were more frequently involved. The app source code analysis allowed us to identify 6 classes of common faults causing specific GUI failures.

On the Resiliency of Static Forwarding Tables

Fast reroute and other forms of immediate failover have long been used to recover from certain classes of failures without invoking the network control plane. While the set of such techniques is growing, the level of resiliency to failures that this approach can provide is not adequately understood. In this paper, we embarked upon a systematic algorithmic study of the resiliency of forwarding tables in a variety of models (i.e., deterministic/probabilistic routing, with packet-header-rewriting, with packet-duplication). Our results show that the resiliency of a routing scheme depends on the “connectivity” k of a network, i.e., the minimum number of link deletions that partition a network. We complement our theoretical result with extensive simulations. We show that resiliency to four simultaneous link failures, with limited path stretch, can be achieved without any packet modification/duplication or randomization. Furthermore, our routing schemes provide resiliency against k - 1 failures, with limited path stretch, by storing log(k) bits in the packet header, with limited packet duplication, or with randomized forwarding technique.

A Unifying Model of Orientation Crowding in Peripheral Vision

Peripheral vision is fundamentally limited not by the visibility of features, but by the spacing between them [1]. When too close together, visual features can become "crowded" and perceptually indistinguishable. Crowding interferes with basic tasks such as letter and face identification and thus informs our understanding of object recognition breakdown in peripheral vision [2]. Multiple proposals have attempted to explain crowding [3], and each is supported by compelling psychophysical and neuroimaging data [4-6] that are incompatible with competing proposals. In general, perceptual failures have variously been attributed to the averaging of nearby visual signals [7-10], confusion between target and distractor elements [11, 12], and a limited resolution of visual spatial attention [13]. Here we introduce a psychophysical paradigm that allows systematic study of crowded perception within the orientation domain, and we present a unifying computational model of crowding phenomena that reconciles conflicting explanations. Our results show that our single measure produces a variety of perceptual errors that are reported across the crowding literature. Critically, a simple model of the responses of populations of orientation-selective visual neurons accurately predicts all perceptual errors. We thus provide a unifying mechanistic explanation for orientation crowding in peripheral vision. Our simple model accounts for several perceptual phenomena produced by crowding of orientation and raises the possibility that multiple classes of object recognition failures in peripheral vision can be accounted for by a single mechanism.

Bringing Modern Unit Testing Techniques to Sensornets

Unit testing, an important facet of software quality assurance, is underappreciated by wireless sensor network (sensornet) developers. This is likely because our tools lag behind the rest of the computing field. As a remedy, we present a new framework that enables modern unit testing techniques in sensornets. Although the framework takes a holistic approach to unit testing, its novelty lies mainly in two aspects. First, to boost test development, it introduces embedded mock modules that automatically abstract out dependencies of tested code. Second, to automate test assessment, it provides embedded code coverage tools that identify untested control flow paths in the code. We demonstrate that in sensornets these features pose unique problems, solving which requires dedicated support from the compiler and operating system. However, the solutions have the potential to offer substantial benefits. In particular, they reduce the unit test development effort by a few factors compared to existing solutions. At the same time, they facilitate obtaining full code coverage, compared to merely 57--72% that can be achieved with integration tests. They also allow for intercepting and reporting many classes of runtime failures, thereby simplifying the diagnosis of software flaws. Finally, they enable fine-grained management of the quality of sensornet software.

Is our heart a well-designed pump? The heart along animal evolution.

A carrier system for gases and nutrients became mandatory when primitive animals grew larger and developed different organs. The first circulatory systems are peristaltic tubes pushing slowly the haemolymph into an open vascular tree without capillaries (worms). Arthropods developed contractile bulges on the abdominal aorta assisted by accessory hearts for wings or legs and by abdominal respiratory motions. Two-chamber heart (atrium and ventricle) appeared among mollusks. Vertebrates have a multi-chamber heart and a closed circulation with capillaries. Their heart has two chambers in fishes, three chambers (two atria and one ventricle) in amphibians and reptiles, and four chambers in birds and mammals. The ventricle of reptiles is partially divided in two cavities by an interventricular septum, leaving only a communication of variable size leading to a variable shunt. Blood pressure increases progressively from 15 mmHg (worms) to 170/70 mmHg (birds) according to the increase in metabolic rate. When systemic pressure exceeds 50 mmHg, a lower pressure system appears for the circulation through gills or lungs in order to improve gas exchange. A four-chamber heart allows a complete separation of systemic and pulmonary circuits. This review describes the circulatory pumping systems used in the different classes of animals, their advantages and failures, and the way they have been modified with evolution.

Hypothesis Testing in Feedforward Networks With Broadcast Failures

Consider a countably infinite set of nodes, which sequentially make decisions between two given hypotheses. Each node takes a measurement of the underlying truth, observes the decisions from some immediate predecessors, and makes a decision between the given hypotheses. We consider two classes of broadcast failures: 1) each node broadcasts a decision to the other nodes, subject to random erasure in the form of a binary erasure channel; 2) each node broadcasts a randomly flipped decision to the other nodes in the form of a binary symmetric channel. We are interested in whether there exists a decision strategy consisting of a sequence of likelihood ratio tests such that the node decisions converge in probability to the underlying truth. In both cases, we show that if each node only learns from a bounded number of immediate predecessors, then there does not exist a decision strategy such that the decisions converge in probability to the underlying truth. However, in case 1, we show that if each node learns from an unboundedly growing number of predecessors, then the decisions converge in probability to the underlying truth, even when the erasure probabilities converge to 1. We also derive the convergence rate of the error probability. In case 2, we show that if each node learns from all of its previous predecessors, then the decisions converge in probability to the underlying truth when the flipping probabilities of the binary symmetric channels are bounded away from 1/2. In the case where the flipping probabilities converge to 1/2, we derive a necessary condition on the convergence rate of the flipping probabilities such that the decisions still converge to the underlying truth. We also explicitly characterize the relationship between the convergence rate of the error probability and the convergence rate of the flipping probabilities.

How to Avoid Future Problems Based on Previous Failure Analyses

All failure analyses have the implicit message that the discovered root cause will somehow be mitigated. A few even suggest the direction one might take for the fix—for example, “Consider substituting Type 316 stainless for Type 304.” This paper emphasizes what might be done to avoid recurrence of the failure. Engineers in managerial positions must lead the corrective action process. Consideration is given to classes of failures for which the fixes can similarly be applied by category. There is little new in this arena of application, but in practice the misapplication of these principles is seen daily. The information presented may provide seasoned engineers some leverage when constrained by budgets and can serve as a tutorial for younger engineers on how to avoid these pitfalls as they take over managerial jobs. The applications considered include low stress–low cycle fatigue, weldments, stress concentrations, and corrosion.

Artificial-Immune-System-Based Detection Scheme for Aircraft Engine Failures

A detection scheme based on the artificial immune system paradigm was developed for specific classes of aircraft jet engine actuator and sensor failures, including throttle, burner fuel-flow valve, variable nozzle-area actuator, variablemixer-area actuator, low-pressure spool-speed sensor, low-pressure turbine exit static-pressure sensor, and mixer pressure-ratio sensor. The NASA Modular Aero-Propulsion System Simulation model was linearized and interfaced with a supersonic fighter aircraft model and a motion-based flight simulator, providing the adequate framework for development and testing. Several engine actuator and sensor failures weremodeled and implemented into this simulation environment. A five-dimensional hyperspace was determined to build the self within the artificial immune system paradigm for detection purposes. The artificial immune system interactive design environment based on evolutionary algorithms developed at West Virginia University was used for data processing, detector generation, and optimization. Flight-simulation data for system development and testing were acquired through experiments in a motion-based flight simulator over extended areas of the flight envelope. The performance of the artificial-immune-system-based detection scheme was evaluated in terms of detection rates and false alarms. Results show that the artificial-immune-system-based approach has excellent potential for the detection of all of the classes of engine failures considered.

Takagi-Sugeno Fuzzy Model-Based Flight Control and Failure Stabilization

This paper presents a Takagi–Sugeno fuzzy model-based approach to modeling, controlling, and stabilizing an aircraft in both normal and damaged conditions. The major contributions of this paper are as follows. First, it introduces the Takagi–Sugeno modeling and the parallel distributed compensation control approach. Then, it demonstrates how the approach can be applied to aircraft in a systematic manner. Finally, the approach is applied to the modified NASA F-15 number 837 nonlinear model, and preliminary simulation results are presented. The benefits of using a Takagi–Sugeno model-based parallel distributed compensation control approach include a systematic controller design procedure, mathematical proof of stability for classes of failures, and intuitive insight into damage effects on an aircraft. This work is the first to present this control approach as applied to aircraft.

System-specific static code analyses: a case study in the complex embedded systems domain

In this paper, we are exploring the approach to utilize system-specific static analyses of code with the goal to improve software quality for specific software systems. Specialized analyses, tailored for a particular system, make it possible to take advantage of system/domain knowledge that is not available to more generic analyses. Furthermore, analyses can be selected and/or developed in order to best meet the challenges and specific issues of the system at hand. As a result, such analyses can be used as a complement to more generic code analysis tools because they are likely to have a better impact on (business) concerns such as improving certain software quality attributes and reducing certain classes of failures. We present a case study of a large, industrial embedded system, giving examples of what kinds of analyses could be realized and demonstrate the feasibility of implementing such analyses. We synthesize lessons learned based on our case study and provide recommendations on how to realize system-specific analyses and how to get them adopted by industry.

The use of reliability‐oriented field feedback information for product design improvement: a case study

AbstractIn the innovative industry, four major trends are found to influence product quality and reliability: the increase in product complexity, the strong pressure on time to market, the increasing global economy, and the decreasing tolerance for quality problems. Thus, it becomes more difficult to anticipate all potential failures during the product development process. In this context, an efficient field feedback process should be in place to react to the unanticipated deviations in product performance. Based on a case study made in an innovative company, this paper shows that the problem is not so much in the information collection as in the inherent quality of the information and in the manner the information is processed. Therefore, a new method, presented in this paper, was developed to classify and prioritize field data and to upgrade it into information that can be used for design improvement according to the dominant classes of failures using the four‐phase roller coaster model. Although this newly generated information is richer than raw field data it is not yet detailed enough to allow direct design optimization. Therefore, a second upgrading stage, based on design of experiments, was developed. It uses a method that combines physics‐of‐failure (bottom‐up) and field information (top‐down). As traditional DoE mainly deals with largely time‐independent quality data obtained during the manufacturing process the approach had to be modified to deal with time‐dependent reliability data. Case study results show that it is a promising approach for characterizing and resolving failure mechanisms also in innovative companies. Copyright © 2008 John Wiley & Sons, Ltd.

Failure detectors for wireless sensor-actuator systems

Wireless sensor-actuator systems (WSAS) offer exciting opportunities for emerging applications by facilitating fine-grained monitoring and control, and dense instrumentation. The large scale of such systems increases the need for such systems to tolerate and cope with failures, in a localized and decentralized manner. We present abstractions for detecting node failures and link failures caused by topology changes in a WSAS. These abstractions were designed and implemented as a set of reusable components in nesC under TinyOS. Results, which demonstrate the performance and viability of the abstractions, based on experiments on an 80 node testbed are presented. In the future, these abstractions can be extended to detect and cope with larger classes of failures in WSAS.

A Methodology for Reliability Analysis in Health Networks

A reliability model for a health care domain based on requirement analysis at the early stage of design of regional health network (RHN) is introduced. RHNs are considered as systems supporting the services provided by health units, hospitals, and the regional authority. Reliability assessment in health care domain constitutes a field-of-quality assessment for RHN. A novel approach for predicting system reliability in the early stage of designing RHN systems is presented in this paper. The uppermost scope is to identify the critical processes of an RHN system prior to its implementation. In the methodology, Unified Modeling Language activity diagrams are used to identify megaprocesses at regional level and the customer behavior model graph (CBMG) to describe the states transitions of the processes. CBMG is annotated with: 1) the reliability of each component state and 2) the transition probabilities between states within the scope of the life cycle of the process. A stochastic reliability model (Markov model) is applied to predict the reliability of the business process as well as to identify the critical states and compare them with other processes to reveal the most critical ones. The ultimate benefit of the applied methodology is the design of more reliable components in an RHN system. The innovation of the approach of reliability modeling lies with the analysis of severity classes of failures and the application of stochastic modeling using discrete-time Markov chain in RHNs.

Comparing reliability and security: Concepts, requirements, and techniques

Design of highly secure systems is increasingly important. This paper explores the relationship between reliability engineering and security engineering for software products. Our work has indicated that designing for reliability and designing for security share many common ideas and, more importantly, techniques that can be leveraged in product development. Security failures are one aspect of product dependability, and while security failures are qualitatively different from other reliability failures in some ways, the two classes of failures have a great deal in common. International Telecommunication Union, Telecommunications Standardization Sector (ITU-T) Recommendation X.805 suggests that availability has a specific meaning for security, namely, that it “ensures that there is no denial of authorized access to network elements, stored information, information flows, services and applications due to events impacting the network.” This definition is very similar to the definition of availability that reliability engineers use in describing functionality. This paper explores the similarities and differences in approaches taken within the communications domain for reliability and security. The three areas that are discussed are terminology, requirements, and common techniques. © 2007 Alcatel-Lucent.