Improving the quality of the driving cycle by processing statistical data of vehicle movement

Abstract

The article describes the process of analyzing and processing statistical data on the movement of freight transport obtained using navigation equipment installed on the vehicles in question. A detailed analysis of the source data, as well as all stages of their processing, was carried out using specialized Python tool packages such as Pandas, Numpy and Sklearn. The effectiveness of the proposed approach to data processing is shown by comparing the driving cycles of vehicles generated on the basis of the original and modified data. The topic of data analysis and processing is not new, but the relevance of the article lies in the successful application of common and effective processing methods to statistical data on the movement of vehicles, work with which is complicated by the fact that they are obtained using widespread, but focused on other tasks equipment. Thus, the work opens up new opportunities for statistical analysis of the movement of freight transport without the need to replace navigation equipment.