Parameter Estimation of a Multistate Model for an Aging Piece of Equipment under Condition-Based Maintenance

• Main Authors: Qihong Duan, Xiang Chen, Dengfu Zhao, Zheng Zhao
• Format: Article
• Language: English
• Published: Hindawi Limited 2012-01-01
• Series: Mathematical Problems in Engineering
• Online Access: http://dx.doi.org/10.1155/2012/347675

We study a multistate model for an aging piece of equipment under condition-based maintenance and apply an expectation maximization algorithm to obtain maximum likelihood estimates of the model parameters. Because of the monitoring discontinuity, we cannot observe any state's duration. The observation consists of the equipment's state at an inspection or right after a repair. Based on a proper construction of stochastic processes involved in the model, calculation of some probabilities and expectations becomes tractable. Using these probabilities and expectations, we can apply an expectation maximization algorithm to estimate the parameters in the model. We carry out simulation studies to test the accuracy and the efficiency of the algorithm.