A condition monitoring approach for real-time monitoring of degrading systems using Kalman filter and logistic regression

E. Skordilis, Ramin Moghaddass

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

We present a new model for reliability analysis that is able to employ condition monitoring data in order to simultaneously monitor the latent degradation level and track failure progress over time. The method presented in this paper is a bridge between Bayesian filtering and classical binary classification, both of which have been employed successfully in various application domains. The Kalman filter is used to model a discrete-time continuous-state degradation process that is hidden and for which only indirect information is available through a multi-dimensional observation process. Logistic regression is then used to connect the latent degradation state with the failure process that is itself a discrete-space stochastic process. We present a closed-form solution for the marginal log-likelihood function and provide formulas for few important reliability measures. A dynamic cost-effective maintenance policy is finally introduced that can employ sensor signals for real-time decision-making. We finally demonstrate the accuracy and usefulness of our framework via numerical experiments.

Original languageEnglish (US)
Pages (from-to)1-18
Number of pages18
JournalInternational Journal of Production Research
DOIs
StateAccepted/In press - Apr 13 2017

Fingerprint

Condition monitoring
Kalman filters
Logistics
Degradation
Monitoring
Reliability analysis
Random processes
Decision making
Sensors
Logistic regression
Kalman filter
Costs
Experiments

Keywords

  • Bayesian filtering
  • classification
  • condition-based maintenance
  • dynamic decision-making
  • Kalman filter
  • logistic regression
  • reliability analysis

ASJC Scopus subject areas

  • Strategy and Management
  • Management Science and Operations Research
  • Industrial and Manufacturing Engineering

Cite this

@article{0671093a7dbb4f8a843d0176e153b34f,
title = "A condition monitoring approach for real-time monitoring of degrading systems using Kalman filter and logistic regression",
abstract = "We present a new model for reliability analysis that is able to employ condition monitoring data in order to simultaneously monitor the latent degradation level and track failure progress over time. The method presented in this paper is a bridge between Bayesian filtering and classical binary classification, both of which have been employed successfully in various application domains. The Kalman filter is used to model a discrete-time continuous-state degradation process that is hidden and for which only indirect information is available through a multi-dimensional observation process. Logistic regression is then used to connect the latent degradation state with the failure process that is itself a discrete-space stochastic process. We present a closed-form solution for the marginal log-likelihood function and provide formulas for few important reliability measures. A dynamic cost-effective maintenance policy is finally introduced that can employ sensor signals for real-time decision-making. We finally demonstrate the accuracy and usefulness of our framework via numerical experiments.",
keywords = "Bayesian filtering, classification, condition-based maintenance, dynamic decision-making, Kalman filter, logistic regression, reliability analysis",
author = "E. Skordilis and Ramin Moghaddass",
year = "2017",
month = "4",
day = "13",
doi = "10.1080/00207543.2017.1308573",
language = "English (US)",
pages = "1--18",
journal = "International Journal of Production Research",
issn = "0020-7543",
publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - A condition monitoring approach for real-time monitoring of degrading systems using Kalman filter and logistic regression

AU - Skordilis, E.

AU - Moghaddass, Ramin

PY - 2017/4/13

Y1 - 2017/4/13

N2 - We present a new model for reliability analysis that is able to employ condition monitoring data in order to simultaneously monitor the latent degradation level and track failure progress over time. The method presented in this paper is a bridge between Bayesian filtering and classical binary classification, both of which have been employed successfully in various application domains. The Kalman filter is used to model a discrete-time continuous-state degradation process that is hidden and for which only indirect information is available through a multi-dimensional observation process. Logistic regression is then used to connect the latent degradation state with the failure process that is itself a discrete-space stochastic process. We present a closed-form solution for the marginal log-likelihood function and provide formulas for few important reliability measures. A dynamic cost-effective maintenance policy is finally introduced that can employ sensor signals for real-time decision-making. We finally demonstrate the accuracy and usefulness of our framework via numerical experiments.

AB - We present a new model for reliability analysis that is able to employ condition monitoring data in order to simultaneously monitor the latent degradation level and track failure progress over time. The method presented in this paper is a bridge between Bayesian filtering and classical binary classification, both of which have been employed successfully in various application domains. The Kalman filter is used to model a discrete-time continuous-state degradation process that is hidden and for which only indirect information is available through a multi-dimensional observation process. Logistic regression is then used to connect the latent degradation state with the failure process that is itself a discrete-space stochastic process. We present a closed-form solution for the marginal log-likelihood function and provide formulas for few important reliability measures. A dynamic cost-effective maintenance policy is finally introduced that can employ sensor signals for real-time decision-making. We finally demonstrate the accuracy and usefulness of our framework via numerical experiments.

KW - Bayesian filtering

KW - classification

KW - condition-based maintenance

KW - dynamic decision-making

KW - Kalman filter

KW - logistic regression

KW - reliability analysis

UR - http://www.scopus.com/inward/record.url?scp=85017407597&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017407597&partnerID=8YFLogxK

U2 - 10.1080/00207543.2017.1308573

DO - 10.1080/00207543.2017.1308573

M3 - Article

AN - SCOPUS:85017407597

SP - 1

EP - 18

JO - International Journal of Production Research

JF - International Journal of Production Research

SN - 0020-7543

ER -