Adaptive data mining as a tool to predict mining machinery and equipment assembly life

Mining industry is one of the most important economic sectors in the modern world. Complex working conditions, high loads and the need for continuous monitoring of equipment technical condition require highly qualified specialists and effective tools to analyze large data volumes. Failure analysis of mining machinery and equipment is one of the important processes to determine and eliminate the causes of failures in order to improve the reliability and safety of machinery and equipment operation. The use of modern methods of statistical data processing makes this process more efficient and accurate. The development of a tool for failure analysis of mining machines and equipment can be very beneficial to mining companies. By analyzing the data on mining machines and equipment failures, identifying the primary causes of failures and providing corrective recommendations, the analysis tool can prevent equipment failures, improve machine safety and performance. The development of this tool requires an interdisciplinary approach as it should be user-friendly and scalable. In this regard, the purpose of the study is to present a creation method of an adaptive tool for the Microsoft Excel-based analysis of mining machine failures. The authors consider the basic operation principles of this tool, its functional composition and application potential under various operating conditions of mining equipment. Much attention is paid to the description of the main operation algorithm of the program, which makes it possible to efficiently process large volumes of data, produce accurate results and display them in the form convenient for reliability level estimation and transition to the forecasting of mining machinery and equipment assembly life. Further improvement of the tool for adaptive analysis of data on mining machine operation, within the framework of this study, can be performed by adding new parameters or automation of the troubleshooting processes using neural networks.