ILWIS objects

Criteria trees

A criteria tree contains the definition of a complicated calculation in which multiple input maps and/or attribute columns are combined, according to certain user-defined rules or criteria, to obtain a final output raster map using Spatial Multi-Criteria Evaluation (SMCE).

Under the main goal of a criteria tree, you can add:

criteria: these rules are either factors or constraints.
groups: as many criteria may be involved, you can create groups under the main goal of a tree to keep an overview of all rules. A group defines an intermediate or a partial goal. Under a group, you can add one or more factors.

A criteria tree as a whole contains all rules you defined; it thus contains the way in which all input data should be standardized, grouped or weighed, to be able to calculate the final output map. The calculated final output map will show the performance of an area for the main goal that was set.

To build a criteria tree, you need to specify:

Names as text	Data and additional information
the name of the main goal	the name of the output raster map that will contain the calculated final result.
the names of the criteria	for each criterion: whether the criterion should be regarded as a factors (benefit or cost) or as a constraints, for each criterion: the input raster map or the attribute column that needs to be used.
the names of groups	optionally for each group: the name of the output raster map that will contain the calculated result for this group, for each group: the names of the factors that will be used in the group, for each factor: the input raster map or the attribute column that needs to be used.

Optionally, you can calculate performance by multiple scenarios or alternatives, in order to select the best scenario or alternative, and to identify poorly performing scenarios or alternatives. You will use:

the same criteria tree, i.e. the same rules for standardization, grouping and weighing of criteria,
for each alternative: other input maps.

Other wording:

The final output raster map is the root of the criteria tree.
An intermediate or partial goal is a branch of the criteria tree.
A criterion or effect is a leaf of the criteria tree.

Names of criteria trees:

In ILWIS 3, object names comply with Windows long file names. Also Universal Naming Convention (UNC) paths are supported. For more information, see How to use long object names.

To create a criteria tree:

in the Main window, choose the Raster Operations, Spatial Multi-Criteria Evaluation command from the Operations menu;
in the Operation-tree, open the Raster Operations item, and select the Spatial Multi-Criteria Evaluation command;
in the Operation-list, double-click the Spatial Multi-Criteria Evaluation command.

The SMCE window opens. In this window the criteria's tree structure can be edited, items can be weighed and standardized and output maps can be generated. For more information, refer to SMCE window : functionality.

To edit a criteria tree:

in the Catalog, double-click a criteria tree, or
select a criteria tree in the Catalog, open the Edit menu in the Main window and choose Edit Object, or
in the Catalog, click a criteria tree with the right mouse button, and select Open from the context-sensitive menu.

The SMCE window appears.

Tip:

To show criteria trees in the Catalog, refer to How to customize the Catalog.

To save a criteria tree:

A criteria tree can be saved by choosing the Save or the Save As commands from the File menu of the SMCE window, or by clicking the Save button in the toolbar of the SMCE window.

Technical information:

A criteria tree only consists of an ASCII object definition file (.SMC). The object definition file stores the number of alternatives, the structure of the tree, the type of the criteria that are used (i.e. groups, factors and/or constraints), references to the input maps, and information on the standardization and weighing methods that are used.

By opening the properties of a criteria tree, you can see the size of the object and the time it was last changed. For more information, see the Criteria tree Properties sheet.