Multicriteria Overlay (Spatial Analyst Tools)
Summary
Combines multiple rasters using a selected multicriteria overlay method, with importance weights applied according to the method.
Usage
Multicriteria Decision Analysis (MCDA) is a structured process for spatial decision-making. First, all criteria are transformed to a common scale. Then the transformed criteria will be weighted and combined by a specified overlay method. The tool performs the overlay stage in this workflow with the selected overlay method defined by the Overlay Method parameter.
The following describes the Overlay Method parameter options and when to use each:
Weighted sum—Use when high values in a criterion can balance out low values. This is the most common method because often you want an overall score by adding weighted criteria together. This is the default method.
Weighted geometric mean—Use when high values in a criterion cannot compensate for low values in other criteria. A single low value will reduce the final suitability because this is a multiplicative method, where the weights are applied as exponents to each criterion value.
Maximum—Use when one high value in a criterion alone is enough to make a location suitable.
Minimum—Use when a single low value in a criterion should disqualify a location.
Weighted overlay—Use to assign each raster a percent influence that adds up to 100. The results are rescaled to an output scale that you defined.
Ordered weighted averaging—Use to explore different decision scenarios by adjusting the Order Weights parameter value. This allows the final result to be more influenced by input criteria that have higher or lower values.
Ideal point solution (TOPSIS)—Use to rank locations by how close they are to a specified ideal and least desired targets. For each location, the method calculates the distance from the ideal and from the least desirable values, then combines these distances into an output cell value.
Ensure that all input rasters have been transformed to a common input scale, such as 1 to 10 or 0 to 1, before using them as input rasters. The exception is when the Overlay Method parameter has been set to the Ideal point solution (TOPSIS) option.
Each transformed raster can be integer or floating-point type. Transforming the criteria to a common scale is necessary because raster values often vary in units, value ranges, and whether higher or lower values indicate greater suitability. This transformation process can be done using the Rescale by Function or Reclassify tools.
When using the Overlay Method parameter's Weighted geometric mean option, the input raster values must be zero or positive. Since the calculation multiplies the values and takes roots, having negative values would produce invalid results.
If a cell has a NoData value in any of the input criteria rasters, the corresponding output cell is assigned NoData.
The Input Rasters parameter's Weight option defines the relative importance of each input criterion raster. In most of the methods, the weights are set to 1 by default. This means that all the inputs are of equal importance.
When using the Overlay Method parameter's Weighted overlay option, weights are specified as Percent, a percentage of influence, and the Percent total must equal 100. By default, the Percent value for the first input raster is 100, and the rest of the input rasters are 0.
For the Weighted overlay method, input raster values will be multiplied by their weights and summed, and then the result will be linearly transformed to the range specified by the From Scale and To Scale parameter values.
The Ordered weighted averaging method combines input raster values based on their ranked order. For each cell, the input raster values will be sorted from highest to lowest. To control how the values are combined, an additional set of weights is assigned to each rank and are then applied sequentially to the ranked values.
Unlike importance weights—which are tied to specific criteria in the Input Rasters parameter—the Order Weights parameter value correspond to the rank positions of the weighted input rasters. The highest value at a cell receives the first rank weight, the next highest receives the second, and so on. By adjusting the Order Weights parameter value, you can make the result more optimistic emphasizing the highest weighted input criteria, more conservative by emphasizing the lowest weighted input criteria, or more balanced.
Order Weights parameter values can be specified manually, or derived using the Orness parameter. The Orness parameter determines the relative influence of higher versus lower ranked criteria on the output.
The Ideal point solution (TOPSIS) option is known as the Technique for Order Preference by Similarity to Ideal Solution. The method calculates the output cell values based on the distance to an ideal positive solution and the distance from the ideal negative solution. Cells that are closer to the ideal positive solution and farther from the ideal negative solution receive higher output values.
When this method is chosen, the Ideal Positive Point Value and Ideal Negative Point Value columns will appear for each input raster. These values represent the best and worst criterion values used for the distance calculation. By default, the Ideal Positive Point Value is the maximum value of the input raster, and the Ideal Negative Point Value is set to the minimum value of each input raster, based on its original value.
For this method, the input criteria do not need to be transformed to a common scale. Instead, the method standardizes each input raster internally. The specified positive and negative ideal values are standardized using the same formula, so they are on the same scale as the standardized input values. The tool then calculates each cell's distance to the positive and negative ideal points using these standardized values.
In the Ideal point solution (TOPSIS) method, the Distance Method parameter provides two options: Euclidean distance and Manhattan distance. Euclidean distance measures the straight-line distance to the ideal point. Manhattan distance measures the total distance by adding the absolute differences across all input rasters.
See Analysis environments and Spatial Analyst for additional details on the geoprocessing environments that apply to this tool.
Parameters
| Label | Explanation | Data type |
|---|---|---|
|
Overlay Method |
Specifies the overlay method that will be used
|
String |
|
Input Rasters |
A list of input rasters will be combined using the specified overlay methods. Value table columns:
|
Value Table |
|
From Scale (Optional) |
The starting value of the output evaluation scale. This parameter is only available if the Overlay Method parameter value is set to the Weighted overlay option. The From Scale parameter value must be less than the To Scale parameter value. The value must be greater than or equal to zero. The default is 1. |
Double |
|
To Scale (Optional) |
The ending value of the output evaluation scale. This parameter is only available if the Overlay Method parameter value is set to the Weighted overlay option. The To Scale parameter value must be less than the From Scale parameter value. The value must be greater than or equal to zero. The default is 10. |
Double |
|
Order Weights Method (Optional) |
Specifies how the Order Weights parameter values will be defined. The weights will be applied after the input rasters have been weighted and ranked. This parameter is only available if you select the Ordered weighted averaging option as the Overlay Method parameter value.
|
String |
|
Order Weights (Optional) |
The set of weights that will be used to combine the input raster values at each input cell location. This parameter is only available when the Overlay Method parameter is set to the Ordered weighted averaging method. The Order Weights parameter values are the weights that will be applied to the weighted input values after they are ranked at each cell (from highest to lowest). These weights control how much influence the largest, middle, or smallest ranked cell values have on the final output values. The order weight values must be greater than or equal to 0. The number of order weights is equal to the number input rasters. The default weight is 1 for each input raster. |
Double |
|
Aggregation Operator (Orness) (Optional) |
Determines how the Order Weights parameter values are derived when the Order Weights Method parameter is set to the Quantifier-guided (Orness) parameter value. The values can range from 0 to 1. The default is 0.5. A value closer to 1 will produce a set of order weights that emphasizes the largest values among the weighted input raster values. A value closer to 0 will produce a set of order weights that emphasizes the smallest weighted input raster value. A value of 0.5 provides a balanced emphasis between large and small values. In this case, the order weights are the same for all input rasters, making the result equivalent to the Weighted sum. |
Double |
|
Distance Method (Optional) |
Specifies how distances will be calculated from each cell to the Ideal Positive Point Value and Ideal Negative Point Value parameters when the Overlay Method parameter is set to the Ideal point solution (TOPSIS) parameter value.
|
String |
|
Output Ideal Positive Raster (Optional) |
The ideal positive raster shows the total distance from each cell to the Ideal Positive Point value when the Order Method parameter is set to Ideal point solution (TOPSIS). The output raster values represent the sum of distances across all input rasters, calculated using the selected Distance Method for each input raster. Lower values indicate cells closer to the ideal positive solution. |
Raster Dataset |
|
Output Ideal Negative Raster (Optional) |
The ideal negative raster shows the total distance from each cell to the Ideal Negative Point Value when the Order Method parameter is set to Ideal point solution (TOPSIS). The output raster values represent the sum of distances across all input rasters, calculated using the selected Distance Method for each input raster. Higher values indicate cells closer to the ideal positive solution. |
Raster Dataset |
Derived output
| Label | Explanation | Data type |
|---|---|---|
|
Output Raster |
The output raster that is the result of applying the selected multicriteria overlay method. It will be of floating-point type. |
Raster Dataset |
Environments
Auto Commit, Cell Size, Cell Size Projection Method, Compression, Current Workspace, Extent, Geographic Transformations, Mask, Output CONFIG Keyword, Output Coordinate System, Parallel Processing Factor, Scratch Workspace, Snap Raster, Tile Size
Licensing information
- Basic: Requires Spatial Analyst
- Standard: Requires Spatial Analyst
- Advanced: Requires Spatial Analyst