Information Technology Reference
In-Depth Information
Designing a semi-quantitative risk graph
Birgit Milius
Institute for Railway Systems Engineering and Tra
c Safety
Technische Universität Braunschweig
Pockelsstraße 3, 38106 Braunschweig
b.milius@tu-braunschweig.de
Abstract. Usually, risk assessments are done qualitatively or quan-
titatively. Neither of these two approaches is satisfactory. In the fu-
ture, risk assessment methods combining the advantages of both ap-
proaches are recommendable. Based on the life cycle process as de-
scribed in DIN EN 50126, the paper discusses guidelines for the con-
struction of semi-quantitative risk assessment methods and presents
a newly developed semi-quantitative risk graph for the application to
the railway sector.
Keywords: Risk Assessment, Safety, Risk Graph
1
Introduction
For changed or new parts of the railway system a risk analysis has to be
performed. In the past, usually quantitative risk analysis methods were used
which are dicult due to not enough data and often time consuming.
A possibility for more effective risk analyses is the use of semi-quantitative
methods, e.g. risk graphs. Semi-quantitative methods are designed based on
a comprehensible mathematical model. In semi-quantitative methods, several
parameters are used for the description of risk. For each parameter a scale
with discrete classes is constructed from which the suitable parameter classes
for the assessment is chosen. From the combination of the selected parameter
classes the requirement class and/or the quantitative safety requirement will
result.
The presented paper discusses guidelines for the construction of semi-
quantitative risk assessment methods and presents a semi-quantitative risk
graph for the application to the railway sector.
2
Classification of Risk Assessment Methods
Risk assessment methods are used for years. According to the classification
developed in [1] methods can be classified according to their parameter assess-
ment, mathematical foundation and tolerability/risk acceptance argument.
