IMPROVING THE RELIABILITY OF TRACTION ROLLING STOCK USING PREDICTIVE ANALYTICS OF ENGINE OIL INDICATORS

Locomotive transport is one of the relevant development areas in the general infrastructure of the sectoral economy. At the moment, a number of tasks and problems related to upgrading existing and the creation of new engine diagnostic systems for this vehicle are being updated. The main purpose of the presented paper is to analyze the possibility of improving the reliability of traction rolling stock by integrating predictive analytics tools for engine oil quality. The authors actualize the tasks related to the need to introduce innovative approaches to analyzing engine oil indicators in real time in order to take appropriate measures to adjust its composition. The paper contains data on the solution to this problem by using predictive analytics tools together with intelligence technologies for monitoring engine oil indicators. The results of the paper describe a possible solution to the problem by using artificial neural networks in determining the current state of the unit, as well as predicting possible changes in the technical condition of the engine relative to operating time. The results also include an algorithmic interpretation of the solution of the initial problem through the introduction of predictive analytics tools. The novelty of the work is in an attempt to comprehensively study and systematize the results regarding the use of predicative analytics to improve the reliability of traction rolling stock by analyzing engine oil indicators. The paper materials are of practical importance, as it is possible to use them to create applied solutions in subsequent research.