Framework for a Predictive Progress Model–case of infrastructure projects

Project progress is an apprehension for every project, as it indicates how the project is likely to meet the associated milestones. Utilizing historical data from archived projects can assist managers in predicting project progress. By leveraging the power of data analytics, this research attempts to highlight data trends based on data collected from 279 infrastructure projects in the UAE. Specifically, this research rigorously analyses the relationships between project budget, duration, and progress using K-means clustering techniques and hypothesis testing. We then provide predictive models using Autoregressive Integrated Moving Average–ARIMùA and Multivariate regression models that allow managers to predict with a 99.15% accuracy the monthly progress of an infrastructure project over the next three months. This research provides project managers with a comprehensive framework that combines data analytics techniques with agility practices to predict short-term project progress to take proactive measures on different influencing factors