The present paper discusses the development of a Markov model for performance evaluation of coal handling unit of a thermal power plant using probabilistic approach. Coal handling unit ensures proper supply of coal for sound functioning of thermal Power Plant. In present paper, the coal handling unit consists of four subsystems with two possible states i.e. working and failed. Failure and repair rates of both subsystems are taken to be constant. After drawing transition diagram, differential equations have been generated. After that, steady state probabilities are determined. Besides, some decision matrices are also developed, which provide various performance levels for different combinations of failure and repair rates of all subsystems. Based upon various performance values obtained in decision matrices and plots of failure rates/ repair rates of various subsystems, performance of each subsystem is analyzed and then maintenance decisions are made for all subsystems. The developed model helps in comparative evaluation of alternative maintenance strategies.