Abstract
AbstractEmergency rescue information is a significant factor affecting emergency decisions in the accident process, and determining the content and importance of this information can greatly assist emergency decisions and improve their effectiveness. For this reason, this paper combines decision‐making trial and evaluation laboratory (DEMATEL) with a backpropagation (BP) neural network and uses the Levenberg‒Marquardt (LM) algorithm to optimize it, builds the LM‐BP‐DEMATEL model, and investigates the contents and importance of emergency rescue information with hazardous chemical spills as an illustrative case. First, the meaning of emergency rescue information was defined using national laws and regulations and academic research. Second, using hazardous chemical spills as an example, 62 hazardous chemical spills were collected, and 7 categories with a total of 32 emergency rescue information were extracted. Third, the constructed LM‐BP‐DEMATEL model was utilized to analyse the importance of 14 types of common emergency rescue information in 62 cases. Last, the centrality and causality of 14 types of emergency rescue information were obtained and then classified into four categories according to centrality and causality. The research results show that deaths (U8) and leakage information (U6) are the most critical emergency rescue information, that more attention should be given to emergency decision‐making and that targeted strategies should be formulated to improve the effect of emergency decisions.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call