ENGINEERING PROJECT RISK MANAGEMENT MODEL

Abstract

This article investigates a comprehensive approach to developing risk management models for engineering and IT projects by integrating standardized methodologies–defined by ISO 31000 and the PMI PMBOK Guide–with modern digital analytical tools. The growing complexity of project environments, rapid technological shifts, and increased systemic uncertainty highlight the urgent need for advanced methods capable of enhancing the accuracy and speed of risk assessment. The study proposes a structured framework that includes the identification, categorization, and both qualitative and quantitative evaluation of risks using a Risk Breakdown Structure (RBS), a probability–impact matrix, and scenario-based analytical techniques such as Bow-tie modelling, Fault Tree Analysis (FTA), and Decision Tree analysis. Particular emphasis is placed on the potential of artificial intelligence–specifically generative models–to support automated diagnostics and reduce subjectivity in expert-based evaluations.
As part of the research, a software prototype titled "Risk-App" was developed to demonstrate the integration of the GPT-API into the risk management workflow. The prototype provides automated risk identification, threat classification, and the generation of response recommendations to support the selection of optimal mitigation strategies. The results indicate that combining traditional risk management practices with intelligent computational tools significantly increases decision-making effectiveness and improves project resilience under multifactor uncertainty. The study concludes that AI-enhanced risk management systems can substantially expand analytical capabilities, accelerate the evaluation process, and improve overall project outcomes by providing real-time, data-driven insights.