• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>System Dynamics Society International Conference; System Dynamics Society Anniversary Celebration; 20070729-0802; 20070729-0802; Boston,MA(US); Boston,MA(US) >A Simulator for Continuous Agent-Based Modelling
【24h】

A Simulator for Continuous Agent-Based Modelling

机译：基于连续Agent的模拟器

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This paper describes a simulation environment that can be used to integrate population-level dynamics with those occurring at an individual, or agent-based, level. The benefit of this approach is that individual agent behaviour may be mapped at a detailed level, using differential equations, and aggregated over the entire population in order to determine population-level dynamics. Furthermore, individual agents can interact with one another, in terms of a social network structure. The environment is firmly grounded in the system dynamics approach, and, unlike conventional agent-based simulation environments, programming is not required in order to specify agent interactions and behaviours. The approach is validated by using a case study based on market dynamics. The overall benefits of the approach are summarised, and future work discussed.
机译：本文介绍了一种模拟环境，可用于将人口级别的动态与个体或基于代理级别的动态进行集成。这种方法的好处是，可以使用微分方程在单个级别上详细映射个体行为，然后将其汇总到整个群体中以确定群体级别的动态。此外，就社交网络结构而言，各个代理可以相互交互。该环境牢固地植根于系统动力学方法中，并且与传统的基于代理的模拟环境不同，不需要编程来指定代理的交互和行为。通过使用基于市场动态的案例研究验证了该方法。总结了该方法的整体优势，并讨论了未来的工作。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号