分布式组织系统的协调行为建模和仿真演化

The Modeling and Simulating Evolution of Coordinating of Distributed Organizations System

针对当前分布式组织研究在内容和方法上存在的局限,考虑分布式组织工作的互依性、位置点的地理分隔、个体的协调沟通和学习等特征,着重分析个体的心理和社会属性,然后借助N人猎鹿演化博弈和行为规划理论构建了分布式组织协调运作的计算模型。通过AnyLogic仿真平台实现该多智能体仿真模型。实验结果表明:(1)自我效能对协调行为的演化和任务完成有着显著影响,而组织的协调性对自我效能的作用很小;(2)组织规模一定的条件下,越不对称的组织构成越有益于组织协调实现和任务的完成;(3)在对称性的组织构成下,组织规模显著影响着协调行为的演化和扩散,但是对自我效能影响很弱;(4)低难度的任务带来较高的收益,提高个体的自我效能,促进组织协调行为的演化,进而有益于组织内部形成良性反馈,而高难度任务可能导致组织内部的恶性循环。

Aiming at the current research limits in contents and methods,this paper builds the computational model of operation in coordinating of distributed organizations based on the N-persons stag hunt game and the theory of behaviorial plan by focusing on mainly in paying attention to the individual psychological and social attributes and the task interdependences and individuals＇ communication and learning characters.We implement this simulation system by virtue of AnyLogic platform.Simulation results show that：（1）self-efficacy significantly affects the coordination evolution and completion of task,while the coordinating has small effect on self-efficacy;（2）under the situation of fixed organizational size,the more asymmetry in organizational compositions,the more benefits the organizational coordination and tasks accomplishments;（3）with the symmetry of organizational compositions,the size remarkably influences the evolution and diffusion of coordination,but has small effect on the individual self-efficacy;（4）the easy tasks can give the higher profits,improve self-efficacy and facilitate the evolution of coordination,thus form the positive feedback in intra-organization,however,the difficult tasks have the negative effect,may incur the negative feedback in intra-organization.