Abstract Centrifugal pumps play a vital role in fluid transport systems across various industrial sectors, where operational efficiency largely depends on parameters such as head and pressure. This study aims to implement a Markov model to analyze and predict transitions in the operational states of a centrifugal pump using historical time-series data. The collected data include inlet and outlet pressure, fluid temperature, flow rate, pump rotational speed, and vibration. After data preprocessing, a discrete Markov chain approach is employed to model state transitions representing different operational conditions. The results indicate that the Markov model effectively predicts variations in head and pressure, as validated against actual performance metrics. These findings support the development of more efficient predictive maintenance systems for centrifugal pumps, offering potential benefits such as improved operational efficiency and reduced long-term maintenance costs.