AUTHOR=Mo Yunyin , Han Qiang , Song Yingying , Li Jingfeng , Zhao Jingjing 

TITLE=Research on prioritization algorithm for multi-hazard warning dissemination

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 13 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2025.1455528

DOI=10.3389/feart.2025.1455528

ISSN=2296-6463

ABSTRACT=The prompt, wide, and effective dissemination of warnings through early warning systems can allows more valuable time for disaster prevention and mitigation. However, due to the limited resources and channel bandwidth of early warning systems, the most urgent warnings may be delayed released when there are multiple pending warnings. To solve this problem, a prioritization algorithm for multi-hazard warning release is proposed. We identified several important attributes of warnings through the analysis of historical data and standardized them, resulting in six priority indicators for early warning issuance. By using entropy method to weight these indicators and combining them into a priority score, the priority of the warning is determined. We conducted retrospective tests on a list of pending warnings that occurred in Guangdong Province on 15 December 2023 and a list of pending warnings that occurred in Jiangxi Province on 23 March 2023. The results showed that this provides an effective method for managing queue systems. In the case of multiple warnings queued to be issued, it can provide an objective and quantitative queuing basis, avoiding biased conclusions drawn from artificial weighted calculation method or single attribute calculations. The algorithm is proved to be indicative when the abnormal warnings occur and improve the timeliness of emergency warning release. In a specific instance, there may be identical values for a certain attribute, resulting in the same score for the indicator determined based on that attribute, and consequently, that indicator does not play a role in the algorithm.we can omit that indicator from the formula to reduce computational load.