RESEARCH OF MACHINE LEARNING METHODS FOR THE ANALYSIS AND FORECASTING OF PROCUREMENT DATA

Abstract

The article is dedicated to the examination of machine learning methods that can be applied for the analysis and forecasting of procurement data. Procurement processes are an essential and integral part of the operation of various business structures and organizations. The establishment and implementation of such processes provide necessary resources, materials, and services to support a company’s operations and contribute to achieving its primary business goals. In the rapidly changing market environment and global economy, most companies face the crucial tasks of analyzing vast volumes of procurement data and predicting the parameters of procurement processes. This is essential for timely responses to evolving market conditions and making effective management decisions. In recent years, with the growth of procurement data volumes, machine learning has become a promising tool for processing, analyzing, interpreting information, and addressing other challenges related to procurement process management. The development and implementation of software systems within logistics management that utilize various machine learning methods for the analysis and forecasting of procurement data can significantly expedite and streamline decision-making processes concerning the acquisition of services and materials. Moreover, it can ensure the high quality of these decisions. The practical value of the research lies in providing recommendations for the utilization of machine learning methods to tackle complex challenges that arise in the management of procurement processes. The article presents an analysis of the advantages and disadvantages of machine learning methods, such as cluster analysis, classification, regression analysis, genetic algorithms, and deep learning. Selecting the appropriate machine learning method for addressing data analysis and forecasting tasks related to procurement processes can lead to high accuracy and effectiveness in analytical results within the intelligent logistics management system.