Optimizing Shortest Job First (SJF) Scheduling through Random Forest Regression for Accurate Job Execution Time Prediction

One of the CPU scheduling methods that is frequently used to reduce waiting time and average execution time is Shortest Job First (SJF). However, this algorithm's accuracy is largelbravy dependent on how well the job execution time is predicted. The purpose of this study is to enhance work execution time estimates by optimizing the SJF algorithm through the use of the Random Forest Regression model. The model in this study is trained using historical job data. The test results demonstrate how Random Forest Regression may be included into SJF to greatly increase system efficiency, especially in terms of throughput and waiting time reduction.