HCC: Advancing the Science of Agent-based Modeling Through Frameworks, Tools, and Pedagogies

在科学实践中，有一场无声的革命正在进行：科学正在变得计算化。丰富的计算能力和最先进的建模软件是科学研究发生巨大变化背后的关键驱动力。但这只是故事的一部分；还有一个根本性的新观点。除了传统上关注系统的宏观属性之外，科学家们正专注于单个原子、细胞或动物的建模。基于这种观点，出现了基于agent的建模（ABM）语言和工具，使研究人员能够以计算格式实现多个相互作用的个体实体的概念模型。计算科学的这一进步将极大地改变科学家和工程师的理论、实践和职业动机。该项目将建立在NetLogo基于代理的建模环境之上，开发框架、工具和课程，进一步将ABM革命带入教室和研究实验室。这项研究不仅适用于工程领域，而且将极大地影响包括自然科学和社会科学在内的许多不同领域的学术和商业世界中基于代理的建模的理论、使用和教学。基于计算主体的科学较少关注纯粹的数值输出，更多关注紧急行为和大规模定性模式，较少关注静态和确定性平衡，更多关注动态和随机过程；因此，它的本质鼓励了复杂系统的建模。这种新的思维方式需要开发计算机体系结构、方法、技术和学习工具来推进和发展这项工作。这个项目将迎接这一挑战。首先，它将开发框架，描述ABM是什么，如何使用它，以及它如何与其他建模和分析技术接口。其次，该项目将创建和改进实现框架的工具，使ABM能够利用多种输入/输出设备，如传感器、执行器和科学成像设备。第三，该项目将生成课程材料，利用这些材料并评估其有效性，以教授新科学家和工程师如何使用ABM。同时进行这项研究可以产生协同效应。通过开发基于主体的建模框架，开发促进基于主体建模的工具，开发用于教授新科学家和工程师的课程，我们的目标是为基于主体的建模创建一个通用的概念和技术基础。这个基础将促进跨学科合作所需的共同理解，并将减少学科之间的重复工作。

There is a silent revolution underway in scientific practice: Science is becoming computational. The abundance of computational power and state-of-the-art modeling software is a crucial driving force behind this sea change in scientific research. But that is only part of the story; there is also a fundamental new perspective. Scientists are focusing on the modeling of individual atoms, cells or animals in addition to the more traditional focus on macroscopic attributes of systems. Out of this perspective have emerged Agent-Based Modeling (ABM) languages and tools, which enable researchers to realize their conceptual models of multiple interacting individual entities in a computational format. This advance in computational science is poised to dramatically change the theory, practice and professional motivation of scientists and engineers. This project will build on the NetLogo agent-based modeling environment to develop frameworks, tools, and curricula to further bring the ABM revolution to classrooms and research labs. This research will be applicable not only within engineering, but will greatly affect the theory, use, and teaching of agent-based modeling in both the academic and business worlds in many disparate fields including the natural and social sciences.Computational agent-based science is less about purely numerical outputs and more about emergent behaviors and qualitative larger-scale patterns, less about static and deterministic equilibria and more about dynamic and stochastic processes; therefore its very nature encourages the modeling of complex systems. This new mindset entails the development of computer architectures, methodologies, technologies and learning tools to carry forward and develop this work. This project will take up that challenge. First, it will develop frameworks describing what ABM is, how to use it, and how it interfaces with other modeling and analysis technologies. Second, this project will create and improve tools that implement the frameworks and enable ABM to take advantage of multiple input/output devices, such as sensors, actuators, and scientific imaging devices. Third, this project will generate course materials, utilize those materials and evaluate their effectiveness to teach new scientists and engineers how to utilize ABM. Doing this research simultaneously affords synergies. Through the development of frameworks to ground agent-based modeling, tools to facilitate agent-based modeling, and curriculum development to teach new scientists and engineers, our goal is to create a common conceptual and technical foundation for agent-based modeling. This foundation will facilitate a shared understanding needed for interdisciplinary collaboration and will reduce duplicated effort between the disciplines.