Premium
Predictive condition‐based maintenance for continuously deteriorating systems
Author(s) -
Lu Susan,
Tu YuChen,
Lu Huitian
Publication year - 2007
Publication title -
quality and reliability engineering international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.913
H-Index - 62
eISSN - 1099-1638
pISSN - 0748-8017
DOI - 10.1002/qre.823
Subject(s) - preventive maintenance , reliability engineering , predictive maintenance , kalman filter , profit (economics) , condition based maintenance , engineering , state space representation , computer science , state space , optimal maintenance , operations research , economics , mathematics , statistics , algorithm , artificial intelligence , microeconomics
This paper addresses a predictive condition‐based maintenance approach based on monitoring, modeling, and predicting a system's deterioration. The system's deterioration is considered as a stochastic dynamic process with continuous degrading. Structural time series, coupled with state‐space modeling and Kalman filtering methods, is adopted for recursively modeling and forecasting the deterioration state at a future time. The probability of a failure is then predicted based on the forecasted deterioration state and a threshold of a failure. Finally, maintenance decisions are made according to the predicted failure probabilities, associated preventive and corrective maintenance cost, and the profit loss due to system performance deterioration. The approach can be applied on‐line to provide economic and preventive maintenance solutions in order to maximize the profit of the ownership of a system. Copyright © 2007 John Wiley & Sons, Ltd.