Tleuova, A. A.Beketova, G. S.2024-12-122024-12-122024Tleuova, A. A. Predictive maintenance using artificial intelligence and machine learning / A. A. Tleuova, G. S. Beketova // Комп’ютерно-інтегровані технології автоматизації технологічних процесів на транспорті та у виробництві : матеріали всеукр. наук.-практ. конф. здобувачів вищ. освіти і молодих учених, 20 листоп. 2024 р. / Харків. нац. автомоб.-дор. ун-т. – Харкiв, 2024. – С. 329–333.https://dspace.khadi.kharkov.ua/handle/123456789/23181Predictive maintenance is a method that helps predict potential equipment failures and avoid unexpected breakdowns. Recently, artificial intelligence (AI) and machine learning (ML) technologies have increasingly been used for predictive maintenance. These technologies allow significant reductions in emergency repair costs and extend the lifespan of equipment. This article explores how these technologies can be applied in Kazakhstan to improve maintenance processes in industries. The aim of the work is to investigate predictive maintenance methods and adapt them to the conditions of Kazakhstan's industry. The following methods are used in the study: Analysis of historical data on failures to predict future breakdowns; Use of sensors to analyze equipment status in real time; Development of hybrid models combining different approaches to improve prediction accuracy. The results of the study showed that using AI and ML for predictive maintenance can help reduce repair and maintenance costs and improve equipment performance. Data analysis from sensors allows timely detection of faults, which helps avoid prolonged downtimes. During the study, failure prediction models were developed, showing prediction accuracy of up to 98.7% with low error rates. These models can be useful not only for large enterprises but also in other sectors of Kazakhstan's economy, such as energy, agriculture, and transportation, which will improve the overall efficiency of the industrial sector.enpredictive maintenanceartificial intelligencemachine learningequipment failuressensorshybrid modelsKazakhstanindustry.Article