Enhancement of short-notice emergency evacuation response during bushfire: a multi-objective epsilon-constraint optimisation approach

A bushfire is an important public health and safety issue in Australia. Late evacuation in bushfire is a crucial stage of emergency response, which requires a quick emergency response to potentially a life threatening condition. Inadequate time to evacuate residents has resulted 119 deaths in the Black Saturday 2009 bushfire in Victoria, which is 68 per cent of total fatalities. Hard constraint of limited evacuation time window (clearance time), uncertainty associated with the direction and intensity of bushfire, road disruption, and elderly and disabled evacuees pose challenges to timely evacuation of people. A decision support system, which is capable of computing timely allocation of resources, is thus required to enhance the capacity of fire services agencies. This paper develops a multi-objective optimisation model for assigning shelters and vehicles and evaluating routing options to transfer late evacuees from assembly points in bushfire affected areas to shelters. Three bushfire scenarios are generated to incorporate constraints of restricted time-window, potential road disruptions, the capacity and availability of rescue vehicles and shelters. The multi-objective problem is solved by Ɛ-constraint technique. The objective functions are simultaneously optimised to maximise the total number of late evacuees by minimum assignment of rescue vehicles and shelters. This model develops a decision support system for fire agencies to minimise resource utilisation and maximise coverage of bushfire affected areas