Information Technology Reference
In-Depth Information
autonomous components assure the safety of operation. This definition is
backed by e.g. the UIC [8]. It has to be shown that the analysis level is the
same level for which a benchmark risk is defined.
Definition of function and hazard:
The term function was not defined as
suitable definitions exist. Hazard was defined as: Hazards in the sense this
work are failures to the unsafe side with a potential of relevant harm.
Type of Result:
The result of the risk assessment is a hazard rate as it
is more flexible than a risk class. A hazard rate can always be transferred
into a risk class but a risk class might lead to more stringent results when
transferred to a hazard rate.
Qualitative and quantitative risk model:
The qualitative risk model is de-
scribing in detail the risk to be assessed. Risk is e.g. a combination of severity
and frequency. The severity is to be measured as harm to people in the train
as this is regarded as a measure of the destructions of the train or its wagons.
Another reason for this approach is that German law focuses on personal risk.
Therefore, looking at damage to persons allows a clear argument. A collective
risk is to be derived as it is assumed that a collective risk for the people in
the train is more significant than an individual risk. It was defined that the
collective to be looked at consists of passengers and workers but not third
parties.
The quantitative risk model for the collective risk was developed based
on the equation for individual risk by Braband [6]:
CRF
F
with
CRF
collective risk of fatality,
HR
hazard rate,
D
hazard duration
time,
E
exposure time,
C
consequence probability and
F
fatalities. For the
quantitative risk model, the equation for
CRF
was reduced to an equation
for risk
R
R
=
HR
·
(
D
+
E
)
·
C
·
F.
Parameters:
Three of the four factors
DE,C
and
F
of the risk model are
used as dynamic parameters for the risk assessment. The fourth factor
HR
is the result of the risk assessment.
=
HR
·
DE
·
C
·
- Harm to people
F
: An indirect approach was taken were the user chooses
the speed at the time of the accident and the type of accident from
a diagram and can directly derive the resulting parameter class. The
diagrams were constructed based on available accident data (more on the
presentation of the parameter severity in e.g. [7]).
- Accident probability
C
: For derailments, the accident probability was
constructed based on the limitations of the infrastructure e.g. relation
between the allowed speed and the maximum speed at which a train
can pass a switch without derailment. For collisions, the parameter was
constructed mainly by combining the rate by which a second train can
be encountered and an estimate of the probability for the train to stop
on time before another standing or driving train.
- Time of exposure and time of the duration of a hazard
DE
: For the time
of exposure this was calculated taking into account basic information
