Measuring railway signals passed at danger

• Authors: George Nikandros;David Tombs
• Venue: SCS '07 Proceedings of the twelfth Australian workshop on Safety critical systems and software and safety-related programmable systems - Volume 86
• Year: 2007

One of the key hazardous events in railways is signals passed at danger (SPAD). SPAD events are relatively common, most with minor consequence. However, SPAD events with catastrophic consequences happen sufficiently often to be a concern within the rail industry. A SPAD event is a precursor to a train collision or derailment; hence there is a strong focus on measuring and reducing their frequency. The railway industry has fostered much research in the role of human factors in SPAD events. Whilst this research has lead to improved selection and training for train drivers and better placement of line-side signals, SPAD events continue to occur and the rail industry is forced to consider expensive technology solutions. Driver error rates are low and may be approaching limits of human performance. While research to improve the reliability of the train driver continues, improvements are hard to measure meaningfully. This paper attempts to quantify driver reliability in relation to SPAD events.