A novel multi-objective optimization framework to allocate support funds for flash flood reduction based on multiple vulnerability assessment

Support funds from national or subnational public spending provide essential assistance for flash floods reduction all around the world. However, how to allocate support funds to different regions with various degrees of disaster is a challenge for decision makers. In this paper, we develop a universal and novel multi-objective framework to solve the problem at multi-space scales. The framework is coupled with the concept of multiple vulnerability and the Fund Allocation Optimization Model (FAOM) developed, taking fully into account the impact of vulnerability on funds allocation. Elite Genetic Algorithm, Technique for Order Preference by Similarity to Ideal Solution and Geographical Detector are applied to solve the FAOM. Based on the case study in Hainan Island, China and the model is compared with the other two fund allocation schemes. Results show that FAOM is able to find a better fund allocation program. As such, it offers a novel approach to solve the problem of funds allocation, considering environmental, social and economic factors. In addition, the further application of the model in progressive recognition and phased support is designed for long-term funding decisions in a particular region, considering the sustainable development of society.