We propose a multi-scale agent-based framework towards understanding and modeling multi-scale interdependent behavioral phenomena. This framework combines the ideas of agent-based modeling with that of hierarchies or levels of organization found in nature and allows for multiple levels in the model to interact at various time scales. We first summarize our rationale for pursuing agent-based models (rather than equation-based models) and then describe the proposed multi-level, multi-scale agent-based modeling framework formally, followed briefly by a discussion of how biological phenomena at different levels could be modeled in the framework. We list some of the requirements and desirable properties for a software simulation tool that can implement such multi-scale, multi-level models, briefly pointing to work in progress on the development of our “tMANS” tool (“the Multi-scale Agent-based Network System”).
@inproceedings{scheutzetal05smc,
title={TMANS - the Multi-Scale Agent-Based Networked Simulation for the Study of Multi-Scale, Multi-Level Biological and Social Phenomena},
author={Scheutz, Matthias and Madey, Greg and Boyd, Sunny},
year={2005},
booktitle={Proceedings of Spring Simulation Multiconference (SMC~05), Agent-Directed Simulation Symposium},
url={https://hrilab.tufts.edu/publications/scheutzetal05smc.pdf}
}