Predictive and Real-Time Analytics for Risk, Cost, and Schedule Control in Healthcare IT Implementation Projects
Kelvin Ebo. Rabbles *
College of Professional Studies, Northeastern University, Massachusetts, USA.
Emmanuel Ivan Ackon
Southern Illinois University Edwardsville, Edwardsville Illinois, USA.
Chioma Emmanuela Ukatu
College of Science, Wayne County Community College, Detroit, Michigan USA.
Isen Bella
College of Professional Studies, Northeastern University, Massachusetts, USA.
Ododoade Adewuyi
College of Professional Studies, Northeastern University, Massachusetts, USA.
*Author to whom correspondence should be addressed.
Abstract
Healthcare information technology (IT) projects are often complex and prone to cost overruns, delays and unforeseen risks, all of which can affect patient care and organisational budgets. This review brings together predictive and real-time analytics models used for risk, cost and schedule control in healthcare IT implementation projects. Literature from project management, healthcare informatics and artificial intelligence was conceptually synthesised from selected studies identified in Scopus, Web of Science and IEEE Xplore between 2019 and 2025. Machine learning models, particularly regression-based, decision-tree and ensemble approaches, are widely used to predict risk assessments and project performance. The key performance metrics reported are Cost Variance (CV), Schedule Variance (SV), Cost Performance Index (CPI), Schedule Performance Index (SPI) and risk probability accuracy (F1-scores: 0.76-0.84). Analytics outputs are operationalised through developed Project Management Offices (PMOs) and governance systems, thereby supporting proactive rather than reactive management. Predictive and real-time analytics offer a paradigm shift for healthcare IT project management by helping organisations move from descriptive to prescriptive insights. However, challenges relating to data interpretability, workflow integration and specialised skill requirements remain barriers to broad uptake. Healthcare organisations should prioritise analytics deployment in high-risk, high-value projects, invest in explainable AI tools and develop hybrid PMO teams with data science expertise. This review is based on selected literature rather than a comprehensive systematic search; quantitative meta-analysis was not possible because of study design heterogeneity, and language bias may exist because searches were limited to English-language publications.
Keywords: Predictive analytics, real-time analytics, healthcare IT, project management, risk management, machine learning, artificial intelligence, explainable AI, earned value management, cost control, schedule control, PMO, governance, Healthcare 4.0, implementation