Abstract: “Government-led, social participation” collaborative disaster relief has become the direction of China's disaster relief system reform. Communication interruption and destruction of transport networks often lead to the delayed rescue. Is the rescue efficiency of government-enterprise cooperation necessarily higher than that of spontaneous rescue in the emergency scenario of delayed rescue? Considering the delayed effect of rescue, a differential game model of government-enterprise disaster relief cooperation is constructed to analyze the influence of rescue delay on government-enterprise disaster relief strategies, and the applicable conditions of government cost subsidy to improve the cooperative disaster relief. Finally, the case of the Ludian earthquake in Yunnan province is verified. The results show that cooperative rescue is more efficient when the delay time is less than a certain threshold. Compared with spontaneous rescue, the government's input increases sharply with the delay time, while the enterprise's input changes little. The enterprise unit income coefficient can be used as an indicator to measure the severity of disasters. The higher the value is, the greater the proportion of government subsidies for disaster relief costs of enterprises.

Key words: disaster relief, differential game, delayed effect, government and enterprise cooperation

摘要： “政府主导、社会参与”的政企协同救灾成为我国救灾体制改革的方向。通讯中断以及运输路网的破坏经常导致救援的延迟,在救援延迟的应急场景下政企协同合作的救援效率一定高于自发救援吗？针对自发救援和协同合作两种救援模式,考虑救援延迟效应,构建了政企救灾合作的微分博弈模型,分析救援延迟对政企救灾策略的影响,以及政府成本补贴对提高协同救灾的适用条件。最后结合云南鲁甸地震的案例加以验证。结果表明,当延迟时间小于某阈值时,协同救援才更有效率。与自发救援相比,协同救援下政府救灾投入量随延迟时间急剧上升,而企业投入量变化较小。企业单位收益系数可作为衡量灾害严重程度的指标,数值越大,政府承担企业的救灾成本补贴比例越大。

关键词: 灾害救援, 微分博弈, 延迟效应, 政企合作