Application of artifikal neural network devices for extrapolating locomotive efficiency indicators

Abstract

The operation of diesel locomotives in different seasons of the year is accompanied by changes in their utilization rates. The efficiency of this process can be assessed by downtime due to unscheduled repairs at any given moment. The disadvantage of the current planned preventive maintenance system for rolling stock is the rather strict regulation of its standards and the lack of effective methods for their operational forecasting, which leads to an inefficient distribution of work on maintenance and repairs and the occurrence of prolonged downtime for unscheduled repairs.