MODELS AND METHODS OF DECISION SUPPORT FORSOFTWARE DEVELOPMENT RISK ASSESSMENT

Abstract

In the modern software development environment, risk management is a critical success factor for projects, especially under conditions of Global Software Development (GSD), where time zone differences, cultural diversity, and communication barriers exist.Traditional risk management approaches, focused on expert assessments or simple probabilistic models, which are insufficient to fully account for the dynamics and complexity of contemporary projects. In response to these challenges, this study proposes the ways of an adaptive Decision Support System (DSS) that integrates modern Artificial Intelligence (AI) approaches, fuzzy logic, machine learning, Bayesian networks, and other methods for both qualitative and quantitative risk analysis.The study is based on an analysis of recent publications and examines the advantages and limitations of separate risk management models, including risk matrices, Bayesian networks, Monte Carlo simulation, fuzzy logic models, and machine learning methods, with a focus on their applicability in different project management contexts. The proposed Decision Support System (DSS), which incorporates a knowledge base, a fuzzy inference engine, and a graphical user interface, has enabled the identification of directions for further improvement to support managerial decision-making in complex project environments. The result of the study is the architecture of a Decision Support System (DSS), which is capable of effectively identifying, assessing, and mitigating risks within complex, dynamic, and distributed teams. The proposed approach is intended to enhance the accuracy of risk prediction, improve the justification of managerial decisions, and contribute to the development of more resilient and productive practices in the field of software development. Suggested architecture provides support for decision-making regarding the formation of an effective team configuration, taking into account risks that impact the execution of a project.