Exploring the limits of train driver reliability

An understanding of train driver reliability can support safety decision making regarding train cab design and train operations. Reliability outputs can be used to more reliably inform safety targets, as it can include any... [ view full abstract ]

An understanding of train driver reliability can support safety decision making regarding train cab design and train operations. Reliability outputs can be used to more reliably inform safety targets, as it can include any increase/decrease in exposure rates. This will also help areas such as bench marking, as it creates a more reliable normaliser. An understanding of train driver reliability can also help to inform incident investigations, by allowing driver performance observed during an incident to be compared against human reliability data. To support industry in this area RSSB developed Railway Action Reliability Assessment (RSSB project T270, published 2012), which describes a method for human reliability assessment and provides generic human performance reliability data. Railway Action Reliability Assessment has been used in a variety of projects in the rail industry, for example it has recently been used to understand signaller performance reliability for an incident investigation undertaken by the RAIB (RAIB report 08/017 - Near miss between a train and a level crossing user at Dock Lane). 

Previous research (“Where is the platform, wrong side door release at stations” Basacik and Gibson, 2015 in Contemporary Ergonomics and Human Factors) has presented driver reliability data on wrong side door release. This paper will provide new data from train companies for other train tasks, including signals passed at danger, station fail to calls and station overruns. The data will be compared to generic performance reliability figures presented in Railway Action Reliability Assessment. The collected data indicate that driver performance provides better reliability than would be predicted using human reliability assessment techniques and previously available data. An initiative within the rail industry relates to collecting better data on the number and locations for red aspects (Stow et al “Estimating the Frequency of Trains Approaching Red Signals – a Key to Improved Understanding of SPAD Risk” IET Intelligent Transport Systems Volume 10, Issue 9, November 2016). The paper will describe how the data on red aspect approach has been used to develop driver reliability estimates for SPADs and how it has been used to inform thinking on SPADs within the GB rail industry. The strengths and weaknesses in the data collection methods will be discussed as well as insights on the underlying issues associated with the data from incident investigations and operation specialists.

We would like the paper to be considered for inclusion in a special edition of Applied Ergonomics.

The submission has industrial application and therefore may be suitable for scheduling in a session focused on the industrial application of human factors/ergonomics.