![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081109150244im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Framework - Modeling Coupling
Research Programs
The MIMS framework relies on the Dynamic Information Architecture System (DIAS) 
, developed by Argonne National Laboratory, for its modeling coupling paradigm and execution management software. Instead of simply representing a collection of models as a network of programs that interact directly with each other, as some other modeling frameworks do, DIAS provides a higher level conceptual structure that promotes interchangeability. The first step in creating a simulation in DIAS is identifying the "domain objects", or things and concepts, that are to be simulated. Examples of domain objects include the atmosphere, a stream, soil, fish, people, and chemicals. For each domain object, the parameters that might be required to describe it are identified. Examples of parameters include atmospheric temperature and a person's age. Each domain object's relevant processes are also identified. Examples of processes include deposition of pollutants from the air to the land, percolation in soil, and death of fish. Identifying domain objects and their parameters and processes provides a common conceptual structure for organizing a simulation. Developing this common conceptual structure can be one of the harder parts of creating a simulation in DIAS. Once the domain objects have been created, models are attached as implementations of the domain objects' processes. For instance, an air quality model might implement the atmospheric deposition process. The models are defined to read and write domain objects' parameters. Thus, models do not know if they are sharing information with other models. This helps minimize the effort required to interchange different models for the same process.
The DIAS software library provides basic templates for domain objects and models and capabilities for constructing interacting sets of models and executing those models in the proper order. The MIMS project is supporting the addition of new capabilities to the DIAS software library. The MIMS framework layers on top of DIAS generic user interfaces, well defined parameter types, functionality that minimizes that programming effort required to bring new models into a MIMS simulation, and additional tools to support modelers.