Asset-specific bayesian diagnostics in mixed contexts (original) (raw)

Improving diagnostic accuracy by blending probabilities: Some initial experiments

Proceedings of the 18th International Workshop on Principles of Diagnosis (DX-07), 2007

Inspired by the impending availability of asset specific data on several US Department of Defense programs, in a previous paper we looked at the possibility that a set of Bayesian diagnostic models constructed from asset specific data would outperform a single Bayesian diagnostic model constructed from all of the data. There were situations where a set of asset-specific classifiers was superior to a single composite classifier but it wasn't universally the case. The hypothesis in this paper is that a blended classifier can be ...

DYNAMIC INTEGRATION OF PROBABILISTIC INFORMATION FOR DIAGNOSTICS AND DECISIONS

The automobile has advanced from a basic means of transportation to a rolling platform that hosts safety, performance, emissions control, and entertainment systems. Due to this increase in complexity, more advanced diagnostic techniques are required. This thesis presents Multi-Modal Diagnostics (MMD), which is a probabilistic approach for diagnosing vehicles and other complex systems. MMD combines model-based diagnostics, Bayesian networks, and statistical decision analysis into a unified probabilistic framework. This thesis introduces the framework and analyzes Being a graduate student at Stanford University has been an amazing experience. I would like to acknowledge those who encouraged me to pursue a Ph.D., those who made it enjoyable, and those who helped me finish. My adviser Prof. Chris Gerdes falls into all of these categories. His passion for exciting research convinced me to remain at my alma mater in order to work with him. He fostered a creative lab environment, which allowed me to take the research in fascinating directions that neither of us foresaw. Over the years he has been not only a teacher and mentor, but also a friend.

Distributional Smoothing in Bayesian Fault Diagnosis

IEEE Transactions on Instrumentation and Measurement, 2009

Previously, we demonstrated the potential value of constructing asset-specific models for fault diagnosis. We also examined the effects of using split probabilities, where prior probabilities come from asset-specific statistics and likelihoods from fleet-wide statistics. In this paper, we build upon that work to examine the efficacy of smoothing probability distributions between asset-specific and fleet-wide distributions to further improve diagnostic accuracy. In the current experiments, we also add environmental differentiation to asset differentiation under the assumption that data are acquired in the context of online health monitoring. We hypothesize that the overall diagnostic accuracy will be increased with the smoothing approach relative to a fleet-wide model or a set of asset-specific models. The hypothesis is largely supported by the results. Future work will concentrate on improving the smoothing mechanism and in the context of small data sets.

Deriving Bayesian Classifiers from Flight Data to Enhance Aircraft Diagnosis Models

ABSTRACT Online fault diagnosis is critical for detecting the onset and hence the mitigation of adverse events that arise in complex systems, such as aircraft and industrial processes. A typical fault diagnosis system consists of:(1) a reference model that provides a mathematical representation for various diagnostic monitors that provide partial evidence towards active failure modes, and (2) a reasoning algorithm that combines set-covering and probabilistic computation to establish fault candidates and their rankings.

1 Novel Classifier Fusion Approaches for Fault Diagnosis in Automotive Systems

2009

Faulty automotive systems significantly degrade the performance and efficiency of vehicles, and oftentimes are the major contributors of vehicle breakdown; they result in large expenditures for repair and maintenance. Therefore, intelligent vehicle health-monitoring schemes are needed for effective fault diagnosis in automotive systems. Previously, we developed a data-driven approach using a data reduction technique, coupled with a variety of classifiers, for fault diagnosis in automotive systems. In this paper, we consider the problem of fusing classifier decisions to reduce diagnostic errors. Specifically, we develop three novel classifier fusion approaches: class-specific Bayesian fusion, joint optimization of fusion center and of individual classifiers, and dynamic fusion. We evaluate the efficacies of these fusion approaches on an automotive engine data. The results demonstrate that the proposed fusion techniques outperform traditional fusion approaches. We also show that learn...

Novel Classifier Fusion Approaches for Fault Diagnosis in Automotive Systems

IEEE Transactions on Instrumentation and Measurement, 2000

Faulty automotive systems significantly degrade the performance and efficiency of vehicles, and oftentimes are the major contributors of vehicle breakdown; they result in large expenditures for repair and maintenance. Therefore, intelligent vehicle health-monitoring schemes are needed for effective fault diagnosis in automotive systems. Previously, we developed a datadriven approach using a data reduction technique, coupled with a variety of classifiers, for fault diagnosis in automotive systems [1]. In this paper, we consider the problem of fusing classifier decisions to reduce diagnostic errors. Specifically, we develop three novel classifier fusion approaches: class-specific Bayesian fusion, joint optimization of fusion center and of individual classifiers, and dynamic fusion. We evaluate the efficacies of these fusion approaches on an automotive engine data. The results demonstrate that dynamic fusion and joint optimization, and class-specific Bayesian fusion outperform traditional fusion approaches. We also show that learning the parameters of individual classifiers as part of the fusion architecture can provide better classification performance.

Detection of Novel Asset Failures in Predictive Maintenance Using Classifier Certainty

2024

Model reliability is one of the main building blocks of trustworthy AI where models are capable of indicating their prediction certainty given the provided input data. Despite the numerous model calibration techniques available to gauge the reliability of trained models, it has been shown that even wellcalibrated classification models cannot properly represent their uncertainty for data points outside the training distribution. Such models, therefore, cannot be used in industrial settings to find previously unidentified failures in an asset, leaving the maintenance crews in the dark. To counter that, we analyze the sequence of the predicted probability values from a condition monitoring classification sub-model of a predictive maintenance solution. We demonstrate that the data points from unknown regions of the asset parameter space can be flagged by simply analyzing such sequences. The flagged failure instances of the asset are consequently sent for thorough inspection to the maintenance crew. Furthermore, we demonstrate on a practical real-world scenario the effectiveness of our proposed method (mean accuracy of 94.35%) by comparing it with models trained only on pre-processed data (mean accuracy of 27.96%) or models based on autoencoder reconstruction error (mean accuracy of 51.09%). We finally show how the predicted probability values from different regions of the parameter space exhibit a higher degree of separability between normal working condition, known and unknown failure modes compared to signal readings of the system or reconstruction errors of an anomaly detection model.

Novel classifier fusion approahces for fault diagnosis in automotive systems

2007 IEEE Autotestcon, 2007

Faulty automotive systems significantly degrade the performance and efficiency of vehicles, and oftentimes are the major contributors of vehicle breakdown; they result in large expenditures for repair and maintenance. Therefore, intelligent vehicle health-monitoring schemes are needed for effective fault diagnosis in automotive systems. Previously, we developed a datadriven approach using a data reduction technique, coupled with a variety of classifiers, for fault diagnosis in automotive systems [1]. In this paper, we consider the problem of fusing classifier decisions to reduce diagnostic errors. Specifically, we develop three novel classifier fusion approaches: class-specific Bayesian fusion, joint optimization of fusion center and of individual classifiers, and dynamic fusion. We evaluate the efficacies of these fusion approaches on an automotive engine data. The results demonstrate that dynamic fusion and joint optimization, and class-specific Bayesian fusion outperform traditional fusion approaches. We also show that learning the parameters of individual classifiers as part of the fusion architecture can provide better classification performance.

Factoring a priori classifier performance into decision fusion

SPIE Proceedings, 2002

In this paper we present methods to enhance the classification rate in decision fusion with partially redundant information by manipulating the input to the fusion scheme using a priori performance information. Intuitively, it seems to make sense to trust a more reliable tool more than a less reliable one without discounting the less reliable one completely. For a multi-class classifier, the reliability per class must be considered. In addition, complete ignorance for any given class must also be factored into the fusion process to ensure that all faults are equally well represented. However, overly trusting the best classifier will not permit the fusion tool to achieve results that rate beyond the best classifiers performance. We assume that the performance of classifiers to be fused is known, and show how to take advantage of this information. In particular, we glean pertinent performance information from the classifier confusion matrices and their cousin, the relevance matrix. We further demonstrate how to integrate a priori performance information within an hierarchical fusion architecture. We investigate several schemes for these operations and discuss the advantages and disadvantages of each. We then apply the concepts introduced to the diagnostic realm where we aggregate the output of several different diagnostic tools. We present results motivated from diagnosing on-board faults in aircraft engines

Factoring a priori classifier performance into decision fusion

Sensor Fusion: Architectures, Algorithms, and Applications VI, 2002

Asset-specific bayesian diagnostics in mixed contexts (original) (raw)

Related papers