A review of evacuation modelling software for emergency management in Australia

Use of community evacuations in Australia as a life-saving strategy for populations threatened by natural hazards such as bushfires, floods and cyclones, has increased in the last 15 years. This increase has followed recommendations for States and emergency agencies to improve evacuation planning and processes from several major inquiries following natural disasters including 2009 Victorian Bushfires Royal Commission [1], 2017 Cyclone Debbie Review [2], Inspector General for Emergency Management (IGEM) Inquiry into the 2019–20 Victorian Fire Season [3], 2020 NSW Bushfire Inquiry [4], and 2020 Royal Commission into National Natural Disaster Arrangements [5]. This period has seen an associated increase in the use of software technology to assist in evacuation decisions. For the purpose of this report, software tools intended for supporting evacuation-related decision-making from planning to response and across different hazards are referred to as Evacuation Simulation Decision Support Tools (ESDSTs).

This research was conducted under a project funded and supported by Natural Hazards Research Australia (NHRA) and sponsored by Department of Energy, Environment and Climate Action (DEECA). The project will operate between 2025-27 with research being conducted by a team from Collaborative Consulting Co and CSIRO. The project seeks to establish a comprehensive, cross-jurisdictional understanding of ESDST use for emergency management in Australia and develop a nationally agreed and prioritised roadmap for ESDST capability development.

The purpose of this review is to document the evacuation modelling approaches currently in use or previously applied within the emergency sector in Australia, providing practitioners with a clear and consolidated picture of existing capability and work to date. This review also serves to align understanding of these tools among stakeholders, given the varying degree of maturity in use of evacuation modelling software across jurisdictions, use cases and hazards. Establishing this shared baseline will support discussions in workshops planned for this project, for identifying user needs and research gaps for improving ESDSTs, and producing a nationally agreed and prioritised roadmap for further development of such technology.

The scope of this report is limited to ESDSTs used by emergency services in Australia for decision support in relation to large-scale community evacuations. Coverage of academic research was restricted to work showing evidence of collaboration with the emergency services. The review does not extend to building evacuations. This review also does not cover data requirements for ESDSTs.

This report primarily focuses on technical reports, government archives, and internal emergency management documentation, and cites peer-reviewed scientific literature only where relevant, such as when readers may seek additional information on specific scientific models or case studies. The analysis also draws on approved information from media releases, public product and project websites, internal reports from emergency agencies, and unpublished work identified through a targeted survey and direct consultation with CSIRO experts.

This research identified 11 ESDSTs that have been used by emergency services in Australia. Five of those (Flood Evacuation Model (FEM), HEC-LifeSim, Probabilistic Tsunami Hazard Assessment (PTHA), Simulations of Emergency Evacuations for Knowledge, Education and Response (SEEKER), and Statewide Analysis of Fire Evacuation Risk (SAFER)) were classed as actively used, given evidence of use in the past five years. Use among jurisdictions varies significantly, with the most use in New South Wales (NSW) and Victoria (VIC), and no identified use in Australian Capital Territory (ACT), Northern Territory (NT), South Australia (SA), and Tasmania (TAS). All 11 models have been used for evacuation planning. In contrast, none of the identified models have been used for response during live emergencies, hindered among other things by a lack of current and accurate real-time data on populations.