The widespread introduction of information technologies in the systems that manage technical fleets, the use of maintenance and repair systems based on risk assessment, is based on the calculation of a large enough number of indicators. Modern locomotives are equipped with systems for monitoring and diagnosing technical condition. Combining these systems with the Internet of Things and Big Data technologies provides an opportunity to use completely new approaches to fleet management. At the initial stage of the construction of such systems, it is necessary to devise criteria that make it possible to automatically determine the technical condition of a locomotive and its components in order to identify the locomotive in the total fleet that requires maintenance or repair.
A procedure has been proposed for calculating the technical condition index of locomotives and their components based on data from monitoring systems. The procedure is based on the formation of latent diagnostic parameters employing the principal component method and on the subsequent calculation of the weight coefficients of these parameters applying the method of hierarchy analysis. The special feature of the proposed procedure is that when calculating the index, those latent diagnostic parameters are used that are derived from the group of control parameters whose weight coefficients are computed using the method of hierarchy analysis without involving experts.
This paper reports the results from calculating the informativeness of the diagnostic parameters of load, loss, input, as well as their weight coefficients. The highest information content, from 0.5 to 0.85, is demonstrated by the load parameter; the smallest (0.05‒0.26) ‒ the input parameter. The average value and the dependences of changes in the technical condition index of a hydraulic transmission during the tests have been determined. Analysis of the technical condition index makes it possible to assess the transmission's response to changes in test modes, the dynamics of changes in losses