Prototype of Intelligent Real-Time Decision Support System Based on Anytime Algorithms and NO-SQL Database

Modern research and development of the creation of intelligent (expert) decision support systems, in particular, real-time (IDSS RT) are directly related to the current situation in the field of artificial intelligence, the problem of modelling plausible reasoning. Availability of similar methods for modelling logic in the IDSS RT for monitoring and managing complex systems and processes of various natures allows rapid diagnosis of a problem situation. It helps decision-makers to find effective management solutions to normalize the problem situation. In several cases, the source data may have a rather complex structure, making it difficult to process it. Thus, document and graph data structures with sufficient data storage flexibility replace the classic strict relational models and tables. The article considers the possibilities of non-relational technologies and anytime algorithms for constructing decision trees based on the applying of the K-th level error level criterion in a combined use modelling IDSS RT for energy facilities.