DATA ANOMALIES IN BUSINESS PROCESS MODELS

Abstract

The article is devoted to the problem of validating business processes when developing solutions based on business process modeling platforms. The article provides a classification of the currently known models of validating data flows in business processes and also considers the types of business processes models: Petri nets, metagraphs, activity-based models, combined models, data flow matrix. The advantages and disadvantages of each model are analyzed, as well as a brief description of the main components of the models. The types of data anomalies and the algorithms used to validate the data flow of a business process are presented. Particular attention is paid to validation methods that use the data flow matrix as the main tool for analyzing data flows. An algorithm for checking anomalies in the absence and redundancy of data using a data flow matrix is given. An example of a model in which a business process works with several domain data models is analyzed. The analyzed example leads to the conclusion that the algorithms built upon the data flow matrix are false when working with several domain data models. The article proposes to find out the possibility of using an ontological model of conceptual objects to identify semantic conflicts between data sets from a variety of domain models.