Business process modeling allows to graphically represent organizational activities and related events. It allows to identify areas for improving the organizational performance, define requirements for software solutions, and, in general, to facilitate communication between IT and business parties within or between different organizations. Therefore, at the stage of representing the activity in the form of a model, it is necessary to understand how likely it is that errors will occur during the implementation of the depicted business process. Thus, this study aims to improve the quality of business process models by solving the problem of predicting the error probability of business process execution. In order to assign error probabilities to each business process model from the training dataset, it is proposed to use one of the complexity metrics – the coefficient of network connectivity. To predict the error probability in business process execution, it is proposed to use the simplest and most intuitive machine learning model – logistic regression. As independent variables, it is proposed to choose the basic metrics of business process modeling – the number of nodes and arcs. Thus, the algorithm for solving the task includes steps related to calculating probabilities for the training data set, preparing training and test sets, determining regression parameters, visualization, and evaluation of training results. For the software that implements the proposed approach, a client-server architecture was chosen due to its flexibility and scalability. When developing software components, the Scikit-Learn machine learning library and the Python programming language were used to build a logistic regression mathematical model. The software tool is implemented as a web application based on MySQL, the Node JS platform, and the Express JS web framework. The quality assessment results of the developed prediction model indicate the suitability of the software tool for solving the problem of predicting the error probability of business process execution.