A volunteer allocation optimization model in response to major natural disasters based on improved Dempster-Shafer theory

Nowadays, factors such as global climate change, environmental damage, and the impact of human activities have led to an increase in natural disasters, and the frequency of natural disaster problems is increasing, which poses a great threat to the safety of people's lives and property. The increased frequency of natural disasters has increased the need to focus on the ability to prevent and respond to disasters. The increased frequency of natural disasters has increased the need to focus on the capacity for disaster prevention and mitigation. All parties need to take appropriate measures to reduce losses and provide assistance after a disaster occurs. Among them, volunteers are a newly emerged important force for rescue, but volunteers have characteristics that normal rescue organizations do not have, such as voluntary and fragmented nature, therefore, volunteer allocation becomes an important issue today. This study tries to construct a volunteer assignment method, which takes into account several factors in the assignment, such as volunteers' own willingness to the disaster site, the competency of volunteers' various abilities, the demand of the disaster site for the task and the time satisfaction of the disaster victims, etc. L-T2FNs are introduced in the assignment process to enhance the degree of certainty; Prospect theory's value function to consider the disaster victims' psychology; This study proposes an algorithm to reduce evidence fusion conflicts. This algorithm uses a differential evolutionary algorithm based on the Dempster-Shafer theory to train the lowest conflict index (determined by K and evidence distance) for BPA fusion. Subsequently, the assignment is demonstrated by using the 7.8 magnitude earthquake in Turkey in 2023 as an arithmetic example to provide a solution to the problem of how to allocate volunteers to the appropriate disaster sites.