No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Assisted Diagnostics Methodology for Complex High-Tech Applications
ResearchGate has not been able to resolve any citations for this publication.
In this paper, we investigate the use of Bayesian networks to construct large-scale diagnostic systems. In particular, we present a novel analytical and experimental approach to developing large-scale Bayesian networks by composi- tion. This compositional approach reects how (often re- dundant) subsystems are architected to form systems such as electrical power systems. We develop Bayesian networks and clique trees representing 24 different electrical power systems, including the real-world electrical power systems ADAPT. ADAPT is representative of electrical power sys- tems deployed in aerospace, and is located at the NASA Ames Research Center. The largest among these 24 Bayesian networks contains over 1,000 random variables. Related work has used Bayesian networks to diagnose specic elec- trical power systems, however we are not aware of previous research that investigates a wide range of distinct electrical power systems as is done in this paper. While we consider diagnosis of power systems specically, we believe this work has application to numerous health management problems, in particular in dependable systems such as aircraft and space- craft. an EPS by means of Bayesian networks. The real-time rea- soning challenge is associated with the embedding of AI components, including diagnostic reasoners, into hard real- time systems. The scalability challenge, which is the main focus of this paper, is concerned with how different EPSs, with varying architectures and components, can be repre- sented in varying Bayesian networks for diagnosis, and how space requirements and computation times vary accordingly. The modelling challenge has been addressed by a high-level specication language from which Bayesian networks are auto-generated; the real-time reasoning challenge by off- line compilation of Bayesian networks into clique trees or arithmetic circuits which are used on-line. The scalability challenge is addressed by means of composition, in other words by considering the subsystems making up a system. For example, a electrical power systems can be made up by power storage and power distribution subsystems. We pro- vide in this paper several novel analytical and experimen- tal results that shed light on large-scale BNs developed for electrical power system diagnosis. The experimental part in- cludes results for Bayesian networks and clique trees repre- senting 24 different electrical power systems, including the Bayesian network and clique tree models of ADAPT. Indication of application status (e.g., feasibility analy- sis, research prototype, operational prototype, deployed application, etc.): We discuss the development of diagnos- tic applications for 24 different electrical power systems, including the Advanced Diagnostics and Prognostics Test- bed (ADAPT) (see also http://ti.arc.nasa.gov/ adapt/). ADAPT, which has capabilities for power gener- ation, power storage, and power distribution, is a fully oper- ational electrical power system that is representative of such systems in aircraft and spacecraft. We have developed a di- agnostic application that is an operational prototype working on real-world data from ADAPT. This paper takes the next step by considering how what we have learned from ADAPT can be applied to electrical power systems that are similar to ADAPT, but of different sizes and structures. Specically, we discuss the composition of electrical power systems from power storage and power distribution subsystems, and how this composition is reected in the Bayesian network and clique tree models of these EPSs.
On-product overlay can be improved through the use of context data from the fab and the scanner. Continuous improvements in lithography and processing performance over the past years have resulted in consequent overlay performance improvement for critical layers. Identification of the remaining factors causing systematic disturbances and inefficiencies will further reduce overlay. By building a context database, mappings between context, fingerprints and alignment & overlay metrology can be learned through techniques from pattern recognition and data mining. We relate structure (patterns') in the metrology data to relevant contextual factors. Once understood, these factors could be moved to the known effects (e.g. the presence of systematic fingerprints from reticle writing error or lens and reticle heating). Hence, we build up a knowledge base of known effects based on data. Outcomes from such an integral (holistic') approach to lithography data analysis may be exploited in a model-based predictive overlay controller that combines feedback and feedforward control [1]. Hence, the available measurements from scanner, fab and metrology equipment are combined to reveal opportunities for further overlay improvement which would otherwise go unnoticed.
A Bayesian network (BN) based fault diagnosis framework for semiconductor etching equipment is presented. Suggested framework contains data preprocessing, data synchronization, time series modeling, and BN inference, and the established BNs show the cause and effect relationship in the equipment module level. Statistically significant state variable identification (SVID) data of etch equipment are preselected using principal component analysis (PCA) and derivative dynamic time warping (DDTW) is employed for data synchronization. Elman's recurrent neural networks (ERNNs) for individual SVID parameters are constructed, and the predicted errors of ERNNs are then used for assigning prior conditional probability in BN inference of the fault diagnosis. For the demonstration of the proposed methodology, 300 mm etch equipment model is reconstructed in subsystem levels, and several fault diagnosis scenarios are considered. BNs for the equipment fault diagnosis consists of three layers of nodes, such as root cause (RC), module (M), and data parameter (DP), and the constructed BN illustrates how the observed fault is related with possible root causes. Four out of five different types of fault scenarios are successfully diagnosed with the proposed inference methodology.
Identification of faults in process systems can be based purely on measurement (e.g. PCA), or can exploit knowledge of process model structure to construct a causal network. This work introduces a method to identify most likely causal network in cases when process model is not known. An incidence matrix, showing location of measurements in the plant network structure, and historical process data are used to identify the optimal causal network structure by means of maximizing Bayesian scores for alternative causal networks. Causal subnetworks, corresponding to subgraphs of the process network, are identified by finding the most probable graph based on highest posterior probability of graph features computed via Markov Chain Monte Carlo simulation. Novel Bayesian contribution indices within the probabilistic graphical network are proposed to identify the potential root-cause variables. Application to Tennessee Eastman Chemical plant demonstrates that the presented method is significantly more accurate than the current methods.
Computers and Chemical Engineering j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c o m p c h e m e n g a b s t r a c t In the design of chemical/energy production systems, a major challenge is to identify the bottleneck issues and improve its sustainability effectively. Due to the multi-dimensional feature of sustainability, how to account for the impacts of various design factors and the cause-and-effect relationships can be very difficult. This paper will present a sustainability root cause analysis method based on the combination of Pareto Analysis and Fishbone diagram. The sustainability of the process is assessed incorporating economic, environmental, societal and efficiency concerns. This methodology is able to help the designers focus the attention on the most important fundamental causes, discover opportunities for sustainability improvement and provide critical guidance to design for sustainability. The efficacy of this methodology will be demonstrated through a case study on a biodiesel production technology.
I discuss an application of a family of Bayesian network models—known as models of independence of causal influence (ICI)—to classification tasks with large numbers of attributes. An example of such a task is categorization of text documents, in which attributes are single words from the documents. The key that enabled application of the ICI models is their compact representation using a hidden variable. The issue of learning these classifiers by a computationally efficient implementation of the EM algorithm is addressed. Special attention is paid to the noisy-or model—probably the best-known example of an ICI model. The classification using the noisy-or model corresponds to a statistical method known as logistic discrimination. The correspondence is described. Tests of the noisy-or classifier on the Reuters data set show that, despite its simplicity, it has a competitive performance. © 2006 Wiley Periodicals, Inc. Int J Int Syst 21: 381–398, 2006.
This paper addresses the challenges inherent in root cause diagnosis of process variations in a production machining environment. We develop and present a process monitoring and diagnosis approach based on a Bayesian belief network for incorporating multiple process metrics from multiple sensor sources in sequential machining operations to identify the root cause of process variations and provide a probabilistic confidence level of the diagnosis. The vast majority of previous work in machining process monitoring has focused on single-operation tool wear monitoring. The focus of the present work is to develop a methodology for diagnosing the root cause of process variations that are often confounded in process monitoring systems, namely workpiece hardness, stock size, and tool wear variations. To achieve this goal, multiple sensor metrics have been identified with statistical correlations to the process faults of interest. Data from multiple sensors on sequential machining operations are then combined through a causal belief network framework to provide a probabilistic diagnosis of the root cause of the process variation.
Probabilistic programming promises to make probabilistic modeling easier by making it possible to create models using the
power of programming languages, and by applying general-purpose algorithms to reason about models. We present a new probabilistic
programming language named Figaro that was designed with practicality and usability in mind. Figaro can represent models naturally
that have been difficult to represent in other languages, such as probabilistic relational models and models with undirected
relationships with arbitrary constraints. An important feature is that the Figaro language and reasoning algorithms are embedded
as a library in Scala. We illustrate the use of Figaro through a case study.
We describe an Early Warning System (EWS) which enables the root cause analysis for initiating quality improvements in the manufacturing shop floor and process engineering departments, at product OEMs as well as their tiered suppliers. The EWS combines the use of custom designed domain ontology of manufacturing processes and failure related knowledge, innovative application of domain knowledge in the form of probability constraints and a novel two step constrained optimization approach to causal network construction. Probabilistic reasoning is the main vehicle for inference from the causal network. This inference engine provides the capability to do a root cause analysis in manufacturing scenarios, and is thus a powerful weapon for an automotive EWS. This technique is widely applicable and can be used in various contexts in the broader manufacturing industry as well.
Combining Evidence using Bayes' Rule
- S D Anderson