The construction industry deals with many risks worldwide, which can cause delays, cost overruns, and inefficiencies. In India, things are even tougher due to rapid urban growth, limited resources, and poor data usage. Most traditional risk management strategies are pretty theoretical and don’t apply in real-life situations. This study aims to create a framework that helps tackle uncertainties in construction risk management by using Big Data and Predictive Analytics. The goals are to understand current risk management practices, pinpoint key factors, evaluate how effective predictive analytics can be, and suggest ways to integrate big data into construction projects. To do this, the research uses the random forest method, which is a machine learning technique that processes real-world data to check predictive capabilities. By focusing on real-world applications, this investigation aims to tackle the limitations of previous studies that have mostly remained theoretical. The results show that predictive analytics can significantly improve how risks are identified and assessed. The framework provides practical insights that can help enhance decision-making and make better use of resources. Recommendations include using real-time data integration tools and encouraging collaboration among stakeholders to manage risks proactively. This research is useful for policymakers, project managers, and industry experts, helping them create more resilient and efficient construction practices. Overall, it contributes to moving construction management from theoretical ideas to practical, data-driven solutions.