Driver Only Operation & CCTV
Railway Safety & Standards Board
background
Railway Group Standards state that a maximum of 4 separate CCTV views can be displayed on a monitor under Driver Only Operation of Passenger trains (DOO(P)). A train operating company had proposed a method whereby up to 6 images would be displayed (maintaining minimum acceptable targets sizes) in order to be able to operate 12 car trains using only two monitors.
RSSB, as the industry safety body, had concerns with this proposal were that drivers might be less reliable at detecting incidents when having to look at more images and that station dwell time might be adversely affected.
project
We were commissioned to undertake a controlled laboratory based experiment where subjects were asked to view sequences of CCTV footage captured from an in-service train which included a number of simulated incident scenarios (targets items). Subjects were asked to determine whether a target item was present, or whether it was safe to move the train, for trains of 4, 6, 8 and 12 lengths. The aim was to determine whether reliability at target detection would significantly decrease and whether task time would significantly increase with more displayed images.
implementation
Analysis of the experimental results found no significant difference in train driver's ability to detect incidents between 4, 6 and 12 car trains.
For 8 car trains a significant difference in incident detection performance was strongly linked to failures to detect single frozen camera views (a CCTV processing system failure). This probably resulted from the fact that the system active indicator does not appear over a camera view in the 8 car layout which may mean that subjects did not necessarily focus on it or attend to it. There is also a possibility that the layout of images for 8 car trains does not help the adoption of an effective systematic search pattern.
However, a key operational findings was that, as one might expect, the more images are presented the more time is required for inspection and therefore the longer the necessary dwell time.
outcomes
A number of recommendations were made including developing effective image scanning protocols and training drivers to use them.
Also suggested were the likely requirements for scanning time to be taken into consideration when planning station dwell times.
This work has been taken forwards on a number of separate projects by specific train operating companies. We have developed some unique knowledge and a now refined methodological approach that can be applied to current and planned dispatch situations.
more information
A full copy of our report can be downloaded from the RSSB website.