MACHINE LEARNING METHOD IN SOFTWARE PROJECT MANAGEMENT

Abstract

Software project management is one of the most important tasks in software engineering, since the success of software development depends on the efficiency of this process. According to stakeholder requirements, software must be delivered on time, within budget, and to all required specifications. Software project management includes key tasks such as estimating the effort required for software development, project scheduling, human resource allocation, risk management, and progress monitoring. Getting these tasks wrong can have serious consequences for software companies, including lost revenue, breach of contract, and even bankruptcy. Managing software projects is extremely complex due to the need to consider multiple business and human factors, as well as conflicting objectives. These challenges are compounded when managing medium to large projects. It is in such conditions that artificial intelligence (AI) can play an important role, supporting project managers in making informed management decisions. Since the advent of artificial intelligence, its role in software engineering has grown significantly. Initial successes have been achieved through the application of metaheuristic algorithms and machine learning to solve the problems of project planning and cost estimation. However, recent advances in language models such as ChatGPT have opened up new possibilities for automating and supporting real-time management decisions. These models are able to analyze large amounts of text data, generate recommendations, and even perform some management tasks automatically, making them a valuable tool for software project managers.

This article explores the implementation of machine learning and artificial intelligence (AI) in software project management, demonstrating how these technologies can significantly enhance the efficiency and accuracy of management processes. The primary focus is on the use of machine learning for automating planning, cost forecasting, resource allocation optimization, and risk management. The article describes how machine learning models, such as language models and metaheuristic algorithms, aid in real-time decision-making, reduce cost overrun risks, and increase team productivity. Case studies illustrate the economic benefits of AI implementation. The article also discusses the challenges associated with integrating machine learning into existing management systems and examines its future development prospects in the context of innovative technologies.