Abstract
In the context of the transposition into Belgian law of the Directive 96/82/CE (Council of the European Union, 1996), also called Seveso II Directive, regional authorities are, in particular, responsible for the inspection of Seveso establishments with regards to external safety (Belgian Official Journal, 2001). Moreover, in Belgium, regional authorities have the responsibility to maintain appropriate distances between establishments covered by the Directive and various areas. To apply this Seveso Directive request (article 12), the Walloon Region developed a methodology to quantify individual risk generated by each Walloon Seveso sites. The Walloon approach selected for the risk assessment and the determination of the “consultation zones” is a probabilistic one with particular assumptions (Delvosalle et al, 2006). This methodology which can be summarized in six steps is based on the use of the DNV software Safeti. Amongst the tools available in this software, the individual risk ranking report permits to determine the composition of risk at specific locations from the result of the modelling. Due to a reorganization of the Walloon regional inspection department, a reflexion was begun to develop, for regional authorities, a more specific inspection method for external safety which must permit to identify the most critical installations for the safety beyond the fences of the plant. This method is similar to a classical risk based inspection (Vianello et al, 2013) method but adapted to Walloon land use planning methodology. In fact, it is based on the result of the modelling of the global risk generated by the plant and the analysis of the individual risk ranking report at specific locations wisely defined by the user. This result allows identifying the installations most contributing to the external risk. However, this method must be used with caution (Bragatto et al, 2012). Indeed, this method permits to identify the most important contributors to the external residual risk by taking into account all safety barriers because it is based on land use planning modelling. However, this prioritisation is valid only whether that all safety barriers operate correctly, are properly maintained and are accurately modelled. So, this individual risk ranking report based on inspection method is an additional tool available for inspection teams but it does not totally replace a classical inspection method and reinforces the interest to inspect all safety barriers which are, of course, a major key for internal and external safety.
