Punctuality and railway maintenance

Abstract: Long and frequent train delays, striking many periodically make headlines. The question is raised: what might be done about it? The on-going deregulation of the Swedish railway, with more actors entering the market, contributes to the question's topicality. In this thesis, problems and symptoms concerning punctuality are described and analysed. Herein, punctuality is defined as the extent to which an event takes place when agreed, for example the agreement between passengers and the company selling the ticket, considering the event that the train arrives at a certain time. The terminology in use on punctuality-related entities is floating. Causes for unpunctuality are related to the infrastructure, locomotives, wagons and personnel. Herein, primarily failures related to infrastructure are discussed. Maintenance, i.e. correcting and preventing failures, together with the track, rolling stock, weather, driver, other personnel and the timetable determine how good the punctuality will be. To the ore and steel transports studied it is more important to limit the lengths of the transport times, rather than having high punctuality, i.e. adherence to timetable. Generally, for freight transports it is difficult to give an estimation of the cost for unpunctuality. To find the root cause of a delay is vital, as is identification of symptoms and triggering events, in order to be able to carry out appropriate actions. Often the information needed to do so is lacking in the Ofelia database for infrastructure failures. Comparisons between regions or over time are complexified by the varying policies on how failure coding is to be performed. To improve punctuality by learning from others is made more difficult by e.g. different countries employing different ways of denoting punctuality. In the transport chain of post in Norway, terms related to punctuality varied even more among stakeholders. They considered being informed on estimated time of arrival as important, but getting this information was, despite this, not among the quantified goals. Few requirements were quantified, despite this often being simple to do. The consequences of failure are important to consider, not only e.g. the number of failures. Studying the variation in transport time on different line sections might help in identifying problems, because coding of delays shorter than five minutes is not carried out in the Swedish railway. Such causes of delay include low contact wire voltage, minor problems with locomotives and slippery rails due to leaves on the track. Disadvantageously, this variation is masked in the case of heavy trains or steep and curvy lines. In a studied maintenance organisation, the indicators of maintenance measured the result of the entire maintenance process. In order to manage each sub process better, it is more advantageous to measure the performance of them individually. Considering the specific sub process that decides which maintenance actions are to be carried out, it only uses expenses for the action itself when deciding, not resulting future costs and punctuality. The indicators focus on measuring past performance, not on facilitating planning for the future. A design of information systems facilitating information usage from different functions of the organisation would facilitate informed decisions. Information concerning technology, traffic and economy need to be integrated.