Title: Data mining in rail transport delay chain analysis

Authors: Jouni Wallander; Miika Mäkitalo

Addresses: Department of Business Information, Management and Logistics, University of Technology, P.O. Box 451, FI-33101 Tampere, Finland. ' M-Files Corporation, Hatanpään valtatie 26, FI-33100 Tampere, Finland

Abstract: Railway systems face the demand for efficient, reliable, and low-cost logistic services. Nevertheless, in many countries, punctuality of rail traffic plummet at a poor level. Even though high quality is pursued, there seems to be a lack of broad understanding when it comes to the concatenation of delays. However, understanding rail traffic delay chains is important for improving the performance of rail transport quality. Our research uses a data-mining approach for analysing rail transport delay chains, with data from passenger train traffic on the Finnish rail network. This study illustrates data mining is a useful tool for identifying and mapping the delay chains. It may be concluded that based on a deeper understanding of the delay concatenation, it is possible to develop rail traffic punctuality and the whole railway system. In medium and long-term planning, data-mining analyses of rail traffic can help to develop a more robust timetable structures, and provide tools for rail network planning.

Keywords: railways; rail systems; rail traffic; punctuality; delay propagation; primary delays; secondary delays; data mining; rail transport; logistics services; delay concatenation; transport quality; transport delay chains; decision support; Finland; rail timetables; timetable structures; rail network planning; transport logistics.

DOI: 10.1504/IJSTL.2012.047492

International Journal of Shipping and Transport Logistics, 2012 Vol.4 No.3, pp.269 - 285

Available online: 24 Jun 2012 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article