| dc.description.abstract | After several road tunnel fires with severe consequences in Europe, the European Union published a directive to ensure the safety of road tunnel users in 2004. The directive called Minimum safety requirements for tunnels in the trans-European road network became a game changer in the field of road tunnel safety. To ensure a certain level of safety, the directive states that where a tunnel has special characteristics, a risk analysis shall be carried out to establish whether additional safety measures and/or supplementary equipment is necessary. This risk analysis shall take into consideration possible accidents, which clearly affect the safety of road users, and the nature and magnitude of their possible consequences. For obvious reasons, fire is a possible event that threaten the safety of road tunnel users.
The purpose of a risk analysis in relation to road tunnel safety, is to provide the decision maker with adequate information about the future risks, and to specify preventive and mitigating measures to reduce the risk to an acceptable level. A simple matter in theory, but a complex situation in the real world. Actors with different risk perspective and different levels of knowledge in both fire dynamics and road tunnels in general, equals varying risk results.
A British consultant company has developed a quantitative risk analysis model called LBAQRAMo, which provide the frequency and consequences of road tunnel fires. This thesis presents basic risk science and investigates how Norwegian quantitative risk analysis is used today, and what value it provides the decisions makers. Data is gathered through literature and interviews with actors in close connection to the field of road tunnel safety. The results are compared with the novel British quantitative risk analysis model, to assess if this is an appropriate tool in the Norwegian context.
The thesis concludes that the connection between internal and external validity of quantitative risk analyses is to some extent connected. However some challenges is present regarding the statistical data used in the models, which are considered too general, affecting the users trust to the output of the models. Local factors contributing to accidents and fires in road tunnels are many, and difficult to model. Risk perspective of both developers and users of these models affect how the value of the models are perceived. The business model associated with the LBAQRAMo seems to be insufficient in a Norwegian context, as the decision maker do not have access to the process of how the risks are derived. | |