The reliability of the machinery greatly affects the long-term potential of the grinding of pepper in Mustika village, Paguyaman district, and Boalemo district. The smoothness of the production process depends heavily on the condition of the machine, and if the machine's reliability is disrupted, then it will affect production. The purpose of this study is to determine the probability of the steady state of the machine and the timing of the maintenance of the grinding machine in the Mustika Village of Boalemo district. The Markov chain is a method used to deal with the purpose, whereas the Markov chain is the method used for predicting future events. The final result was an ergodic transition chance matrix, resulting in an estimated best maintenance time of 28 days of use with a steady state chance of 62.27\% of the machine being in good condition, 27.8\% in mild damage, and 9.93\% in severe damage.