The theoretical basis of the choice of new locomotives for Ukraine in the post-war period

In the case of the research of promising locomotives, we are dealing with a complex event – "choosing a locomotive for implementation". To effectively solve this problem, it is suggested to decompose this event. Therefore, the purpose of this work is to develop a methodology for modeling the evaluation process according to objective criteria of various options of new traction rolling stock. The Saaty method has been developed by transforming the hierarchy into an artificial neural network. The training of this network occurs automatically when analyzing the matrices of pairwise comparisons, and at the output we have a generalized criterion – the rating of the locomotive R, the value of which varies from 0 (the worst indicator) to 1. This allowed, unlike the existing approach, not to compare locomotives by compiling a matrix of comparisons at the last stage. Instead, a matrix of comparisons of the most important criteria by which traction rolling stock is evaluated has been compiled. The developed method has the ability to support various strategies for the operation of the locomotive park. This is implemented at the stage of drawing up the second-level criteria comparison matrix. Depending on the tasks facing the railways, it is also possible to adjust the degree of preference of one criterion over another. This provides even greater flexibility in using the proposed method.