关键词: 学习行为, 团队绩效, 自我管理团队, 智能体, 虚拟企业

Abstract: To explore the influence of team learning behavior on team task process, an approach of team learning behavior analysis based on multi-agent simulation was put forward. In the approach, self-managed team in virtual enterprise was taken as the example to examine the effect. After an analysis on self-managed team's double-level learning behavior, team members and tasks were both defined as agents. Based on definitions of concepts as the error rates of self-estimation ability, estimation of others'ability and rules of role changing, a multi-agent simulation model was constructed and implemented into MSSSTB system with Microsoft Visual Basic 6.0. The model was validated on simulation analysis of a practical problem. The simulation results showed that double-level learning behavior was more helpful for promoting efficiency of team task processing than single-level learning behavior. And the effectiveness of MSSSTB system was verified in the analysis of team learning behavior.

Key words: learning behavior, team performance, self-managed team, agents, virtual enterprise