Planning and evaluating project management are key parts of project performance that should not be overlooked. It is difficult to succeed at project management unless you have a realistic and logical plan in place. This paper provides a comprehensive overview of papers on the application of machine learning in software project management, covering a wide range of topics. Apart from that, this study examines machine learning, software project management, and methodologies. Papers in the first category are the results of software project management studies or surveys. Papers in the third category are based on machine-learning methods and strategies applied to projects; studies on the phases and tests that are the parameters used in machine-learning management; and final classes of study results, contribution of studies to production, and promotion of machine-learning project prediction. Our work also provides a larger perspective and context, which could be useful for future project risk management research, among other things. To summarize, we have demonstrated that project risk assessment using machine learning is more effective in minimizing project losses, increasing the likelihood of project success, providing an alternative method for efficiently reducing project failure probabilities, increasing the output ratio for growth, and facilitating accuracy-based analysis of software fault prediction.