FAME

FAME is designed to combine high computational speed with ease of use for developing and deploying agent-based model simulations. Its high execution speed is achieved through the library FAME-Core. This Java package is specifically targeted at supporting simulations in the domain of energy systems analysis. FAME offers parallelised simulation execution without any extra effort from modellers. Thus, any model developed with FAME can be run both in single-core and multi-core mode. FAME applications were deployed on various computer and operation systems, ranging from small desktop computers to large-scale high-performance computing clusters.

FAME supports fast and easy model development by encapsulating common tasks, e.g., reading input data, collecting output data, scheduling agent actions etc. Its flexible contracting system enables adapting agent behaviour through model parameterisation. Thus, model users may modify model behaviour without the need for code changes. FAME also comprises the Python toolbox FAME-Io that facilitates fast and consistent model configuration and also converts model outputs. Its graphical user interface FAME-Gui provides for easy and convenient ways of parameterising models.

Using FAME, model developers may focus on model development only and are finally freed from burdens that usually accompany the creation of parallelised scientific simulation software. All FAME software packages are available open-source.

Check out the documentation.

URL: https://gitlab.com/fame-framework

Contact: fame@dlr.de