Abstract: Maintaining a reasonable reserve of emergency materials in advance is an effective way in response to sudden disasters. But in reality, it is difficult to meet the surge demand in post disaster period only by relying on the government's reserve materials, and suppliers are required to join the emergency material reserve. However, due to the differences of social environment (e.g.disaster probability, social vulnerability, population density) among regions, the emergency material reserves requirements are also different among regions. For example: i)Material reserves should prioritize covering areas with a high probability of disaster occurrence; ii)Due to the difference of social vulnerability in different regions under the same disaster level, emergency material reserves should give priority to cover areas with low vulnerability; iii)The number of people affected by the same level of disaster varies, which requires that population density also be taken into account when preparing emergency supplies. Therefore, it is valuable to study the government-enterprises joint reservation of emergency relief from the perspective of social environment.
In recent years, disaster relief related problems have received extensive attention and in-depth research from both domestic and foreign scholar. The existing literature basically takes the government as the decision maker and applies the operational research method to optimize the facility location, reserve quantity and distribution plan of emergency materials. Most of these literatures are based on P-median, P-center, Set covering, location-allocation, location-routing and other classical models, starting from the simplest static problems and further expanding to stochastic and dynamic problems, which provide theoretical reference for this paper. However,there are still some limitations: i)Existing studies mainly consider the manageable factors such as cost, fund, demand and capacity, while ignoring the differences of social environment; ii)Existing studies only focus on individual dimensions of the demand characteristics, but are lack of the comprehensive consideration of demand characteristics including “diversity”, “modularity”, “timeliness difference” and “fairness” in disaster relief.
Therefore, different from existing papers, we propose a novel bi-objective mixed integer model which not only takes account the social environment(disaster occurrence probability, social vulnerability, population density), but also comprehensively considers the demand characteristics including modularity, timeliness, and fairness requirements of multi-material reserves. The objective function of the model is to minimize disaster risk and system cost at the same time. The variables to be determined include the location plan of emergency facilities, the selection plan of suppliers, the quantity of material reserve, and the distribution plan of emergency materials after disaster. To solve the bi-objective model, a ε-constraint method is applied to get the Pareto solution set.
Finally, we apply our model in a real-case study. A simulation analysis is carried out on the background of the Sichuan earthquake disaster, and the results show: i)The optimal decision of relief allocation is influenced by both social environmental factors and storage requirements; ii)The two goals of risk and cost are contradictory, so it is necessary to set a reasonable budget to achieve maximum emergency efficiency; iii)The marginal utility of unit emergency material reserve cost increases with the increase of disaster level; iv)Increasing the number of suppliers participating in the relief storage process can not only reduce disaster risks, but also reduce costs; v)Although increasing fairness requirements will lead to a decline in emergency reserve efficiency, stronger fairness requirements can reduce the fluctuation range of costs and risk targets, and have higher robustness.

Key words: emergency, cooperation of government and supplier, social environment, material reserve

摘要： 应急物资政企联合配置是提高灾后应急能力及降低成本的有效途径。现实中,应急物资配置需因地制宜,综合考虑灾害发生率、社会脆弱性等社会环境因素,以及充分考虑资源需求特征。为此,本文提出一类全新的社会环境视角下的应急物资政企联合配置问题,综合考虑了社会环境差异对物资储备的影响,全面分析了多物资储备的模块化、时效性、公平性要求,继而基于风险、成本双重目标构建一类实证分析、优化决策联动的物资配置模型,并以四川地震灾害为背景进行仿真分析。结果表明:社会环境因素、多物资储备的模块化、公平性等要求等都会影响应急物资配置的最优决策;增加物资储备环节的供应商参与数量,不但可降低灾害风险,还可降低成本;单位应急物资配置成本的边际效用随灾害等级增加而变大;提高资源分配公平性要求可减小成本和风险目标的波动范围并提高稳定性。

关键词: 应急, 政企联合, 社会环境, 物资配置