Overview

DynaMind is a light weight open source GIS workflow/simulation engine for developers and researchers to make development and sharing of models that work with rich interlinked GIS data sets easier.

Key features of DynaMind are:

  • Interactive workflow engine that allows GIS models to be combined to describe complex processes
  • Standardised and easy to use interfaces building on the GDAL library for C++ and Python to develop GIS models as DynaMind Modules that can be shared with others
  • Support for working with interlinked GIS and non-GIS data.

DynaMind follows a modular design concepts with DynaMind being the heart and sole of the workflow/simulation engine. DynaMind provides the workflow logic and module handling, which can be accessed through following C++ or Python interfaces:

  1. Simulation interface to create, modify and execute simulations
  2. Module interface that enables users to developed their own modules through an easy to use database interface building on in the GDAL library.

To provide a ready to use environment the DynaMind-Toolbox combines several tools building on DynaMind including:

_images/overview.png

Simulations

The figure below gives an overview of the supported features and describes a typical use case.

_images/workflow.png

In a typical simulation a series of modules are combined to define a workflow.

  • Data Import: Firstly, a simulation requires data, often from multiple sources. The standard modules provide therefore a set of modules building on the GDAL library supporting a variety of standard GIS exchange formats (see here) as well as non GIS data.
  • Interlinked Data Set: An interlinked data set provides the foundation for complex simulations. The Standard Modules provide several modules to spatially or non-spatially link different data sets to an interlinked data model that reflects this interrelationships.
  • Manipulate Data: To manipulate the interlinked data set DynaMind provides a number of generic modules that allow to derive attributes from the interlinked data (see here). Combining several of this modules allows complex processes to be described.
  • Grouping of Modules: Models can be combined to a module Group. Such a group can be executed repeatedly. Similar to a for loop, groups can be used to iteratively manipulate data to describe dynamic processes.
  • Export and Publish Results To publish the resulting data sets the Standard Modules provide modules to export data into common GIS exchange formats or GIS databases.