More and more companies are influenced by the rapid development of technology (Industry 4.0/5.0 concept), are embracing digital transformation processes. The introduction of information systems makes it possible to accumulate a large amount of data about the company’s activities. Study of such information expands the opportunities for applying a data-driven approach to business process management (BPM). Processing and studying data from event logs using process mining methods make it possible to build digital models of business processes which turn out to be a useful source of information when carrying out analysis, modeling and reengineering within the framework of the process approach. In this paper, we develop a method for building a business process model based on a hidden Markov model, taking into account the restrictions imposed by the subject area. The use of a hidden Markov model allows us to use the apparatus of probability theory and mathematical statistics to analyze business processes, as well as to solve classification and clustering problems. This article describes the capabilities of a data-driven approach to business process management and demonstrates examples of the practical application of the method to solve business challenges: drawing a dependency graph that can be used to identify discrepancies between actual and expected execution, as well as a method for predicting the outcome of a business process based on the sequence of observed events.