Machine Learning is on the rise and is transforming industries across the board, from climate forecasting to stock price evaluation. In this study, we explore the use of machine learning in real-time scheduling algorithms for cluster environments. Using the “GWA-T-4 Auver Grid” dataset, we predict burst times of processes with an accuracy of over 87%. We then compare the performance of the FCFS and SJF scheduling algorithms using these predictions, and find that while SJF performs better, it is better suited for short processes, while FCFS is better for longer ones. Our results provide insight into the potential of machine learning in the realm of real-time scheduling algorithms for cluster environments.