Author(s)

A. Yamamura, M. Koresawa, S. Adachi & N. Tomii

Abstract

In urban railway lines, it is usually observed that once a small delay happens, it widely spreads out to many other trains. There is a compelling need to identify small delays that very often happen and influence widely and to take an effective countermeasure to decrease those delays by adjusting timetables for example. In this paper, we introduce an algorithm to identify delays that very often happen and influence widely from daily train traffic record data. The key idea of the algorithm is to seek for the causes of delays by tracing back a critical path on a PERT network constructed from train traffic record data. We show the details of the algorithm together with our experience of its application to real world data. Keywords: delay, urban railways, PERT, critical path, train traffic records.

Keywords

delay, urban railways, PERT, critical path, train traffic records