Development of a data quality management methodology based on logical-semantic digital double

Data quality is a key issue in decision support systems. The urgency of the problem is due to the fact that the introduction of corporate information systems, digital technologies and digital twins requires strict synchronization and updating of a large amount of data. Management decisions depend on big data. The use of data containing one or more logical errors can lead to incorrect decisions and large losses for the enterprise. Aim. The aim of the study is to develop a classification of data that takes into account the production profile, types of data in production, the formal structure of business processes for each type of data based on the proposed logical-semantic architecture of the digital twin for data quality control. The theoretical and methodological basis of the study are the works of domestic and foreign scientists on the problems of managing digital platforms and digital transformation and digital twins of the enterprise. Results. In the course of the study, the classification of the main logical-semantic factors of poor-quality data provision was generalized and systematized, and a methodology for assessing data quality was presented. Business processes for collecting, storing and processing are formalized for each type of data and one responsible person is identified for the relevant business processes, as well as competency requirements for individual responsible persons and a tool for data quality control are developed. As a result, an approach based on the logical-semantic architecture of the digital twin for data quality control is proposed. Conclusion. In modern realities, logical filtering of a huge amount of redundant data in decision support systems (DSS) is necessary. Digital twins (DTs) in decision support systems can provide high quality of initial digital data for PPR. This article presents an integrated approach to data quality management based on their logical-semantic digital twins when creating an effective decision support system using the example of oil and gas companies. Data quality management based on the use of logical-semantic digital twins ensures the usefulness of digital data.