ABMS范例

中文字数:16025,中文页数:12  英文字数:8735页数:16

部分组成整体
 基于Agent的建模和仿真（ABMS）是以“许多系统的或组织的整体是比系统的每个部分的总和还要大”的理念为基础的。为了管理这样的系统，我们必须把整个系统理解为一些相互之间有交互的组件的系列。每一个这样的组件必须都有其自己的规则和职责。可能有些组件比其它的更通用更重要些，但是没有任何的组件能够完全控制整个系统的行为。所有的组件们都或多或少地影响最后的（仿真）结果。所以，这样的一系列组件，我们称之为“复杂的适应性系统”（CAS）。这样的系统的一个症状，正是整体的结果大于个体组件的结果之和！而这样的症状叫做“紧急行为”。CAS常规来说能展现出紧急行为。而要管理好一个CAS系统，需要很好地抓住紧急行为。ABMS将这个基础性的观点结合了经过论证的，高度成功的，诸如离散事件仿真、面向对象编程等技术，找到了一条探索战略性的、技巧性的和可操作性的事务解决方案的崭新路线。
 与那些传统的系统建模技术相比，ABMS的差别在哪里？比如拿ABMS与这些相比：数据建模、风险分析、系统优化、动态系统、标准化仿真、抑或是传统的事件仿真，差别何在？上述的每个方案都很实用，而且在过去的几十年里，它们每个都已经被证明是有实际价值：可谓各有独到之处。然而，这些系统仿真技术唯独无法在先进的事务问题中保证它们的精确性。更重要的是，它们在处理高度非线形的交互工作中显得有些力不从心，然而这些通常都会引发紧急事件的问题，却和当今的工业和政府是息息相关的。最后，上述的每个仿真技术的优缺点有将在第5章被一一提及。

Parts Make the Whole
 Agent-based modeling and simulation(ABMS) is founded on the notion that the whole of many systems or organizations is greater than the simple sum of its constituent parts. To manage such systems, the systems or organizations must be understood as collections of interacting components. Each of these components has its own rules and responsibilities. Some components may be more influential than others, but none completely controls the behavior of a complete system. All of the components contribute to the results in large or small ways. Such a collection of components is said to be a complex adaptive system(CAS).System reactions where the complete results are more than the sum of the individual components’ outcomes are called emergent behavior. CAS regularly exhibit emergent behavior. Managing a CAS requires a good grasp of emergent behavior. ABMS combines this fundamental insight with proven, highly successful techniques such as discrete-event simulation and object-oriented programming to produce a new way to discover strategic, tactical, and operational business solutions.
 How does ABMS compare to traditional system-modeling techniques? Specifically, how does ABMS compare to statistical modeling, risk analysis, optimization, systems dynamics, standard participatory simulation, or traditional event simulation? Each of these approaches is useful in its own right and each has proven its value over the years. However, these system-modeling techniques alone are often not adequate to address many of today’s business questions. In particular, they tend to have difficultly capturing the highly nonlinear interactions that are common in the problems addressed by industry and government. These nonlinear interactions regularly combine to produce emergent behavior. Ultimately each of these approaches has strengths and weaknesses that are considered in chapter.

相关论文