Abstract: To improve the utilization efficiency of big data command and decision-making information during the process of major petrochemical fires, this paper proposes a commanding information classification method of petrochemical fires to meet the needs of different decision-making levels of firefighting agents. Firstly, based on case studies and the characteristics of emergency commanding of major petrochemical fires, the key situational requirements and information sets of different agent command level are determined, and an information classification index is put forward to meet the situational requirements of each commanding agent. Secondly, based on game theory, a classification algorithm for major petrochemical fire situation information is proposed. By calculating the importance of key situation information indicators, the importance classification of situation information for different agent command levels is achieved. Finally, the calculation method proposed in this article is validated through an example. The research results show that the calculation results obtained based on this method are basically consistent with actual cases, and the conclusions can provide support for fire rescue teams to improve the utilization efficiency of command and decision-making information for major petrochemical fires and promote the construction of intelligent emergency response systems for chemical fires.

Key words: petrochemical fire, firefighting and rescue, situational information, Game theory, importance classification