Cluster Analysis as a Basis for the Development of an Application Assessing the Reliability of Transport Infrastructure

Abstract

Increased demand for transport services and increased mobility of citizens can lead to a reduction in the level of reliability of transport systems. This in turn increases the demand for ways to assess the reliability of road infrastructure by both transport companies and individual users. The article presents the substantive basis of an application used to determine the reliability of transport infrastructure. Our approach was based on grouping information into clusters – based on the author's proprietary clustering method. Its basis is the detailed analysis of the road infrastructure in terms of errors occurring on it, divided into conceptual, design and operational errors. The methodology consists of three stages of clustering (1) creating a database of sections with assigned errors, (2) determining the initial clusters, (3) creating a final database of clusters, and then assessing the reliability of the road infrastructure of the transport system on their basis. The assumption is that the application will remain open-ended – i.e. the database will be developed by users. The proposed methodology was verified on the example of the selected route in Poland (between Kalisz and Szczecin). Based on the results obtained during the experiment on the selected route, errors in the road infrastructure were determined. This, in turn, allowed us to find that there are a number of errors in the road infrastructure, including errors with a high frequency of occurrence i.e. the so-called permanent errors, which further confirms the need to create an application to assess the reliability of the road infrastructure.