Page 147 - 360.revista de Alta Velocidad - Nº 5

Probabilistic Safety Analysis of High Speed and Conventional Railway Lines

Table 6 Circumstances for incidents associated with variables incident for cases speed-m, speed-b and speed-mb showing the
associated ENSI and corresponding percentages.

Table 7 Circumstances for incidents associated with variables incident for cases speed-m, speed-b and speed-mb showing the
Probabilities and % ENSI values.

Table 7 is very similar, but now we include a column with the probability for each incident
severity and the corresponding ENSI percentage. It is interesting to see the probabilities of
minor, medium and severe incidents for each case. As expected minor incidents occur at the
speed limit sign more frequent than medium and these more frequently than severe incidents.

4. Conclusions

Bayesian network models provide an important tool to perform a probabilistic safety assessment
of railway lines, and are more powerful than the commonly used fault and event trees, especially
when common causes are present.

The proposed model reproduces all the variables involved in the problem, including their
qualitative dependencies and the quantification of the associated conditional probabilities.

In particular, human error must be carefully considered in an integrated form, that is, considering

International Congress on High-speed Rail: Technologies and Long Term Impacts - Ciudad Real (Spain) - 25th anniversary Madrid-Sevilla corridor 145