Skip to content
Bjarne Bergquist
View original picture , opens in new tab/window

Researchers will develop an experimental methodology for railway maintenance

Published: 14 July 2020

Every year, wear and tear on the railroad causes enormous costs, not only in the form of repair costs but also due to delays in train traffic. Now, researchers at Luleå University of Technology will develop models to support the choice of maintenance measures on the railway.

“Railway maintenance is costly and will often lead to considerable traffic disturbances. If our project is successful, the industry will have a method to create better knowledge about which maintenance methods that they can apply. For instance, which methods are most effective and efficient, least costly and which bring the slightest traffic disruptions”, says Bjarne Bergquist, project manager as well as Professor of Quality Technology and Logistics

A Model for effect relationship

The effect of a maintenance measure and how quickly the railway deteriorates after maintenance depends, among other things, on the type of maintenance performed, environmental conditions and maintenance equipment, but also how much and how often the railway is exposed to stress. The overall goal of the research project is to develop a model for mapping the effect relationship between action and long-term results.

“The industry as a whole lacks a validated approach to developing these relationships. This is a problem, not only in Sweden because there is little research in the field”, says Bjarne Bergquist.

Two major case studies

The research group, which, in addition to Bjarne Bergquist, consists of three other researchers in Quality Technology and Logistics from Luleå University of Technology, will collaborate with engineers from the Swedish Transport Administration and BDX (who are responsible for the maintenance of freight traffic in parts of Norrbotten) in two major case studies. In these case studies, they will try to develop cause-and-effect relationships for maintenance measures. At the same time, they will study the process of developing these relationships. The aim is to analyse, formalise and improve the method as well as the experiments used to develop cause-and-effect relationships. The researchers will both observe and participate; while observing how the engineers work, they will advise on how and what kind of measurement data should be collected. They will also point out which actors in the work process who need certain information and when in the process they need it. Furthermore, the researchers will actively participate in the choice of the experimental design as well as explain the advantages and disadvantages of the different experimental designs.

In one case study, the project group will look at the parameters that control, among other things, energy requirements and maintenance needs of the switch-and-crossing heating system. Heat is used to prevent the switch from freezing in the winter, and it is very costly to heat the switch.

The second case study is about track tamping. The stress invoked by trains and seasonal changes eventually lead to incorrect track geometry, which in the worst case can cause a derailment. The problem can be countered, for example, by replacing the ballast - the top layer of the embankment on which rail grinders rest. This is a costly measure to be considered as a reinvestment rather than maintenance. A more common approach is to lift the rail, level the ballast and then put the rail back in its correct position.

Finding the most effective method

By experimental tests of the different methods available for generating heat for switches and crossings and tamping, the researchers hope they will be able to determine which methods are most effective and efficient.

“I have worked for a long time with railway maintenance by looking at measurement data that others have collected. Working with secondary data poses many difficulties because they were collected for other purposes. This will be the first time in railroad contexts that I will be involved in producing primary data experimentally. Thus, we researchers can control the data collection so that it meets its intended purpose. This means, among other things, that we will be able to use calibrated and validated measurement methods. I look forward to it because it is in the development of data that we have our perhaps most important research expertise,” says Bjarne Bergquist.

The project will last for three years during the period 1 August 2020 to 31 July 2023. The Swedish Transport Administration has granted the project research funding of just under SEK 3 million. The research group includes, in addition to Bjarne Bergquist, Erik Vanhatalo, Peter Söderholm and Francesca Capaci.