CONCEPTUAL MODELING OF ADAPTIVE UNCERTAINTY MANAGEMENT IN INFORMATION TECHNOLOGY PROJECTS

Abstract

The paper addresses the timely scientific and practical problem of enhancing the effectiveness of managing information technology projects in high-uncertainty environments with rapidly evolving requirements. The study aims to develop and substantiate a conceptual model of adaptive management that formalizes decision-making processes in response to unforeseen events (unknown unknowns). The methodological foundation combines agile development practices, risk management theory, and the Unified Modeling Language (UML). The limitations of classical deterministic project management approaches are demonstrated in the context of a stochastic execution environment typical for IT projects. Based on a systems analysis, a comprehensive model of adaptive uncertainty management is proposed. This conceptual model of adaptive uncertainty management is based on the hypothesis that effective adaptation is only possible when managerial interventions are synchronized across four levels: process, structural, role-based, and object-oriented. The process view (Activity Diagram) algorithmizes the cycle of processing uncertainty signals by integrating feedback loops and continuous updates to the project knowledge base. The structural view (Class Diagram) specifies the domain ontology and establishes logical relationships among risk sources, adaptation mechanisms, and response strategies. The role-based view (Swimlane Diagram) details interaction among the development team, architect, and product owner during technical validation, including the use of exploratory tasks. The behavioral view (State Diagram) models the task life cycle by introducing specific blocking and exploration states, enabling the quantitative assessment of uncertainty's impact on flow metrics. The model enhances project risk management by embedding the «Agile Spike» instrument into a formalized decision-making framework, thereby treating technical experiments not as costs but as investments in reducing project entropy. The practical contribution lies in a set of reusable visual patterns suitable for implementation in project management information systems, supporting faster responses to uncertainty incidents and higher-quality architectural decisions.