AI-Augmented Project and Program Management: Predictive Analytics for Risk, Cost and Schedule Control

Abstract

The increasing complexity and inherent uncertainty of modern projects have necessitated sophisticated analytical tools to support the consolidation of project and program management decision-making. This paper shows a systematic literature review that analyses the use of artificial intelligence (AI) -enhanced predictive analytics in improving risk, cost, and schedule control in construction, information technology, and healthcare project management. The systematic review identifies the gaps in the current state of knowledge through three research questions: (i) the current application of AI and machine-learning methods to project risk analysis, cost estimation and schedule forecasting; (ii) the success of AI-based tools to enhance project performance outcomes, and (iii) the data, organisational, trust, and ethical issues in the context of AI integration, especially hybrid human-AI models. Results have shown that highly developed AI models, such as neural networks, ensemble learning algorithms, probabilistic and Bayesian models, and natural-language processing, are significantly more accurate in prediction compared to traditional deterministic models. The AI-enhanced tools help to detect cost overruns, schedule slippage, and emergent risks earlier and provoke more proactive and informed managerial interventions. Nevertheless, the review also shows that the advantages of AI are very dependent on the quality of data, the interpretability of the model, and organisational preparedness. Experience always suggests that hybrid models that integrate AI-based insights together with expert judgment and conventional methods of assessing risk, including Monte Carlo simulation, are the most efficient and reliable ones. The paper concludes that AI is mostly a decision-supporting and diagnostic accelerant, but not a substitute for project managers. This review can help academia and practice by synthesising the latest empirical and theoretical literature to discuss how AI-enhanced predictive analytics can provide sustainable changes in project performance and governance under which conditions exist.

Ähnliche Arbeiten