| 摘要: |
| 目的 模拟预测和分析作战中的战斗减员总数、时空分布以及伤员构成分布。方法 使用系统动力学方法构建作战过程模拟和战斗减员预计模型，通过智能体建模方法从战斗减员预计模型中导入宏观减员数据，并对减员数据进行拆分和按特定比例赋予战伤信息。结果 基于系统动力学的战斗减员预计模型能够结合具体作战任务，分析作战影响因素、双方武器杀伤性能和防护水平。对交战过程构建因果回路和存量流量关系，将作战中红蓝双方各类目标毁伤程度转换为减员数据。提取战斗减员预计模型得出的宏观减员数据，通过构建作战目标毁伤程度与各类战伤之间的对应关系，实现对每一个战伤减员个体的具体伤情赋值和模拟，完成从减员流到伤员流的转换。结论 所建立的基于系统动力学的减员预计模型和基于智能体的伤员发生模拟模型，能够科学测算作战中的减员时空分布和减员结构。 |
| 关键词: 减员预计 伤员发生 系统动力学 多智能体 |
| DOI：10.16781/j.0258-879x.2018.05.0510 |
| 投稿时间：2017-12-26修订日期：2018-02-06 |
| 基金项目:军队后勤科研计划重大项目（AWS14L012）. |
|
| A simulation research on casualty prediction based on system dynamics and agent-based modeling |
| PENG Bo,ZHANG Wen-qin,DU Guo-fu,XU Lei* |
(Institute of Health Service and Transfusion Medicine, Academy of Military Medical Sciences, Academy of Military Sciences PLA China, Beijing 100850, China
*Corresponding author) |
| Abstract: |
| Objective To simulate, predict and analyze the total number, spatial and temporal distribution, and proportion and composition distribution of combat casualties. Methods System dynamics was used to construct an combat process simulation and casualty prediction model. Agent-based modeling was used to import macro casualty data from the prediction model, split the casualty data and assign the combat injury information in a specific proportion. Results The casualty prediction model based on system dynamics could integrate with specific operational mission and analyze the combat influencing factors, weapon destruction performance, and level of protection in both Red and Blue sides. The casual-effect loop and the stock-flow model were constructed on combat process. The degree of damage to the target of the two sides in the battle was transformed to casualty data. We extracted the macro casualty data from the combat casualty prediction model. Through constructing the corresponding relationship between the destruction degree of operational objectives and war wound information of all kinds, we assigned and simulated the traumatic condition of each individual casualty and completed the conversion from casualty to wounded flow. Conclusion Constructed casualty prediction model based on system dynamics and the casualty generating model based on agent can scientifically calculate the spatial, temporal distribution and proportion and composition of casualties. |
| Key words: casualty prediction casualty generation system dynamics multi-agent |
