A model-based diagnosis problem occurs when an observation is inconsistent with the assumption that diagnosed system not faulty. The task of a engine to compute diagnoses, which are assumptions on health components in explain observation. In this paper, we extend Reiter's well-known theory by exploiting duality relation between conflicts and diagnoses. This means hitting set conflicts, but conflict also We use property interleave search for diagnoses conflicts: can guide diagnosis, computed...

Bayesian belief networks are being increasingly used as a knowledge representation for reasoning under uncertainty. Some researchers have questioned the practicality of obtaining numerical probabilities with sufficient precision to create large-scale applications. In this work, we investigate how precise need be by measuring imprecision in affects diagnostic performance. We conducted series experiments on set real-world medical diagnosis liver and bile disease. examined effects performance...

A standardized 'stress' was applied to groups of normotensive and hypertensive subjects. Systolic diastolic blood pressures rose in both but were significantly greater the than group. Urinary catecholamines with stress a similar extent groups. There no evidence that patients sustained hypertension have an increased production either at rest or under 'stress'.

Recent research has found that diagnostic performance with Bayesian belief networks is often surprisingly insensitive to imprecision in the numerical probabilities. For example, authors have recently completed an extensive study which they applied random noise probabilities a set of for medical diagnosis, subsets CPCS network, subset QMR (Quick Medical Reference) focused on liver and bile diseases. The terms average assigned actual diseases showed small sensitivity even large amounts noise....

The aim of a modern Building Automation System (BAS) is to enhance interactive control strategies for energy efficiency and user comfort. In this context, we develop novel algorithm that uses stochastic building occupancy model improve mean while minimizing expected discomfort. We compare by simulation our Stochastic Model Predictive Control (SMPC) strategy the standard heating method empirically demonstrate 4.3% reduction in use 38.3%

We propose a StochAstic Fault diagnosis AlgoRIthm, called SAFARI, which trades off guarantees of computing minimal diagnoses for computational efficiency. empirically demonstrate, using the 74XXX and ISCAS-85 suites benchmark combinatorial circuits, that SAFARI achieves several orders-of-magnitude speedup over two well-known deterministic algorithms, CDA* HA*, multiple-fault diagnoses; further, can compute range HA* cannot. also prove is optimal propositional fault models, such as...

The genetic code appears to be optimized in its robustness missense errors and frameshift errors. In addition, the is near-optimal terms of ability carry information addition sequences encoded proteins. As evolution has no foresight, optimality modern suggests that it evolved from less optimal variants. length codons also optimal, as three minimal nucleotide combination can encode twenty standard amino acids. apparent impossibility transitions between codon sizes a discontinuous manner...

Detection and identification of failures is a critical task in the automatic control large complex systems. In realm discrete event systems, Sampath et al. (1995, 1996) proposed new approach to failure diagnosis that models logical behavior considered system terms state machines produces an extended observer called diagnoser for computing diagnoses. We extend this timed systems whose temporal are modeled by framework Brandin Wonham (1994). use simple real-world factory conveyor example...

Cyberattacks against Industrial Control Systems (ICS) can have harmful physical impacts. Investigating such attacks be difficult, as evidence could lost to damage. This is especially true with stealthy ; i.e., that evade detection. In this paper, we aim engineer Forensic Readiness (FR) in safety-critical, geographically distributed ICS, by proactively collecting potential of attacks. The collection all data generated an ICS at times infeasible due the large volume data. Hence, our approach...

Accurate Throughput Prediction (TP) represents a real challenge for reliable adaptive streaming in challenging mediums, such as cellular networks. State-of-the-art solutions adopt Deep Learning (DL) models to improve TP accuracy various multimedia systems. This paper illustrates that designing blackbox engines depend solely on the model’s capacity and power of learning does not achieve consistent across all throughput ranges. Additionally, we propose MATURE, novel multi-stage DL-based model...

Computing diagnoses in domains with continuously changing data is difficult but essential aspect of solving many problems. To address this task, a dynamic influence diagram (ID) construction and updating system (DYNASTY) its application to constructing decision-theoretic model diagnose acute abdominal pain, which domain the findings evolve during diagnostic process, are described. For that evolves over time, DYNASTY constructs parsimonious ID then dynamically updates ID, rather than new...

This article describes the impact that autonomous vehicle operation will have on a range of domains and how formal methods can guarantee safe those vehicles. The verification systems (e.g., autopilot for commercial aircraft) was traditionally done using precise models software based methods. However, increasing use machine-learning-based subsystems perception) requires different tools. We describe are used today they be adapted future

Model-based diagnostic reasoning often leads to a large number of hypotheses. The set diagnoses can be reduced by taking into account extra observations (passive monitoring), measuring additional variables (probing) or executing tests (sequential diagnosis/test sequencing). In this paper we combine the above approaches with techniques from Automated Test Pattern Generation (ATPG) and Model-Based Diagnosis (MBD) framework called FRACTAL (FRamework for ACtive Testing ALgorithms). Apart inputs...

This paper presents a comparison between two model based diagnostics methodologies that can be used to detect and diagnose various faults occur in Air Handling Units. The process from development inference is highlighted with emphasis on the requirements for implementing successful diagnosis solution. Comparative results of both an air handling unit are presented thoroughly discussed using as benchmark rule-based approach known air-handling performance assessment rule-set.

We describe a new paradigm for implementing inference in belief networks, which consists of two steps: (1) compiling network into an arithmetic expression called Query DAG (Q-DAG); and (2) answering queries using simple evaluation algorithm. Each node Q-DAG represents numeric operation, number, or symbol evidence. leaf the answer to query, that is, probability some event interest. It appears Q-DAGs can be generated any standard algorithms exact networks (we show how they clustering...

We describe an approach to automate the dynamic computation of optimal control/reconfiguration actions that can achieve pre-specified control objectives. This approach, based on model-based diagnostic representations and algorithms, integrates diagnostics reconfiguration for discrete event systems using a single modeling mechanism suite algorithms. When system functionality degrades (i.e., failures occur in systems), algorithm will isolate most likely failures, then generate least-cost...

We propose a novel framework for Model-Based Diagnosis (MBD) that uses active testing to decrease the diagnostic uncertainty. This is called LYDIA-NG and combines several diagnostic, simulation, active-testing algorithms. have illustrated workings of by building LYDIA-NG-based decision support system Gravity field steady-state Ocean Circulation Explorer (GOCE) satellite. paper discusses model GOCE Electrical Power System (EPS), algorithms diagnosis disambiguation, experiments performed with...