Abstarct: Proper selection of drill bits is considered as an important issue in well planning. Identification of sedimentary rocks before well drilling plays crucial role in choosing the drilling bit. Markov chain probability method is presented as one of the powerful methods for identifying lithological units, which is baxsed on the calculation of the transition probability matrix or transition matrix. Markov chain is a probable pattern that introduces a specific type of dependency and is used as a model for probable prediction. In this paper, modeling by Markov chain in MATLAB software on Dashtak Formation, which has a sequence of rock faces and is considered as one of the most important formations in southern Iran, has done. Then, considering the prediction in the field of lithology type in the next laxyers by Markov chain and input data, a suitable bit is suggested. This process was carried out in two vertical sections in two well where the thicknesses of Dashtak Formation are 963 meters and 1410 meters. The results show that Markov chain model in well 1 and well 2, 88.54% and 75.46% were in accordance with the output state of the data in reality, respectively.