Evolution of emergency preparedness : a case study of two Norwegian road tunnels
Utgivningsinformation: Borås : RISE Research Institutes of Sweden AB, 2023Beskrivning: s. 439-455Ämnen: Onlineresurser: I: Proceedings from the Tenth International Symposium on Tunnel Safety and Security, Stavanger, Norway, April 26-28, 2023Abstrakt: Tunnels are complex sociotechnical systems, involving several stakeholders, e.g., road authorities/owners, road users and the emergency services. A large fire in a tunnel is a potentially high consequence event, which received increased attention after 1999 due to the catastrophic tunnel fire incidents in central Europe. The implementation of European Commission Directive 2004/54/EC in member states led to considerable improvements of the technical infrastructure in tunnels. However, article 15 of the Directive focuses on learning from incidents and sharing the learning points among member countries on a bi-annual basis. Systematic accident reporting and accident investigation are thereby required. In this paper, we focus on two Norwegian tunnels, Gudvanga and Oslofjord, both single tube, with bi-directional traffic and longitudinal ventilation. Between 2011 and 2021, three fire incidents in each tunnel occurred, involving heavy vehicles. The evolution of emergency preparedness in each tunnel is investigated over time. Emergency preparedness was seen as a construct involving technical, operational and organizational measures, thus making changes traceable.The single and double loop theory of organizational learning was used to identify learning processes improving emergency preparedness in the tunnels. Two double loop learning processes were identified regarding safety in Oslofjord tunnel. Context dependent fire ventilation strategy, based on where in the tunnel the fire is located as well as the number and position of road users provides the best prerequisites for successful responses, with minimal smoke exposure for both tunnel users and infrastructure. Additionally, exposed users are provided with shelters, offering temporary protection under TCC supervision, until rescue resources arrive. Several single loop learning processes are identified in both tunnels, while remaining challenging issues are highlighted.Tunnels are complex sociotechnical systems, involving several stakeholders, e.g., road authorities/owners, road users and the emergency services. A large fire in a tunnel is a potentially high consequence event, which received increased attention after 1999 due to the catastrophic tunnel fire incidents in central Europe. The implementation of European Commission Directive 2004/54/EC in member states led to considerable improvements of the technical infrastructure in tunnels. However, article 15 of the Directive focuses on learning from incidents and sharing the learning points among member countries on a bi-annual basis. Systematic accident reporting and accident investigation are thereby required. In this paper, we focus on two Norwegian tunnels, Gudvanga and Oslofjord, both single tube, with bi-directional traffic and longitudinal ventilation. Between 2011 and 2021, three fire incidents in each tunnel occurred, involving heavy vehicles. The evolution of emergency preparedness in each tunnel is investigated over time. Emergency preparedness was seen as a construct involving technical, operational and organizational measures, thus making changes traceable.The single and double loop theory of organizational learning was used to identify learning processes improving emergency preparedness in the tunnels. Two double loop learning processes were identified regarding safety in Oslofjord tunnel. Context dependent fire ventilation strategy, based on where in the tunnel the fire is located as well as the number and position of road users provides the best prerequisites for successful responses, with minimal smoke exposure for both tunnel users and infrastructure. Additionally, exposed users are provided with shelters, offering temporary protection under TCC supervision, until rescue resources arrive. Several single loop learning processes are identified in both tunnels, while remaining challenging issues are highlighted.