- Decision Wizard - develop single or multi-objective multi-criteria evaluation models. The Wizard facilitates use of the modules WEIGHT, MCE, RANK and MOLA. Decision rules are recorded at each step and may be modified at any time.
- Compute a best-fit set of weights for factor images used in a multi-criteria evaluation through a pairwise comparison analysis. Employs the Analytical Hierarchy Process (AHP) with information on consensus and with procedures for resolving conflict.
- Calculate a multi-criteria evaluation by means of a Boolean combination, weighted linear combination or ordered weighted average using constraint and factor images.
- Rank order the cells in a raster image. The procedure is used extensively in optimization problems such as with RECLASS for single objective decisions and MOLA for multi-objective decisions.
- Allocate pixels to multiple objectives based on an iterative conflict resolution decision heuristic.
- Evaluate the fuzzy set membership values (possibilities) of data cells on any of three membership functions: sigmoidal, j-shaped and linear. A user-defined function capability is also available. Monotonically increasing, monotonically decreasing, symmetric and asymmetric variants are supported. Other Fuzzy Set operations such as CON (concentration), DIL (dilution), AND and OR are covered by the modules TRANSFORM and OVERLAY.
- Calculate a relative frequency image of non-zero values over multiple images.
- Resolve conflicts between competing objectives by means of a multi-dimensional choice procedure.
- Evaluate the probability of pixel values exceeding or being exceeded by a user-defined threshold based on the stated RMS error for the input map.
- Evaluate Bayes' Theorem. Multiple evidence maps are permitted so long as they are conditionally independent. Prior probabilities may vary continuously over space.
- Dempster-Shafer Weight-of-Evidence procedure.
- Produce an error matrix analysis of categorical map data compared to ground truth information. Tabulate errors of omission and commission, marginal and total error, and selected confidence intervals. Per-category Kappa Index of Agreement figures are also provided.
 |
|
| IDRISI includes an unparalleled suite of tools for multi-objective/multi-criteria decision support. Multiple suitability problems are defined in terms of factors, constraints, factor weights and risk-taking strategy. Then the composite suitability images for each objective are used together to best allocate areas to each objective. Rather than one-shot black-box solutions, this approach provides decision makers with understandable and defensible methods that can be iteratively improved. Here, industrial and commercial suitability are modeled.
| |  |
|
|
| This screen shot illustrates some of the data products developed for a study on the effects of sea level rise on a coastal rice-producing area in Vietnam. Part of the study included prediction of future sea level and inundated areas. This portion of the study explicitly incorporated uncertainty in the elevation data as well as the projected sea level rise. The probability of inundation image (upper left) was then thresholded based upon an acceptable level of risk. The multi-criteria analysis was a prediction model of where people might relocate after inundation. Thus a predicted landcover image was produced.
| |  |
|
|
| The Order Weighted Averaging (OWA) module provides an interesting alternative to the commonly-used linear weighted combination approach to aggregation of multiple criteria. By varying the importance of the factors in particular order positions, one can adjust the levels of tradeoff between factors and risk aversion in the solution incorporated into the final model. The images show multiple outputs using the same factors and factor weights, but differing order weights for a suitability analysis near Nakuru, Kenya.
| |  |
|
|
| The next two screen shots illustrate the se of the BELIEF module for developing a "weight-of-evidence" model of land degradation. In this first image, we see the degree to which the evidence supports the hypothesis of land degradation (belief). Data is for a portion of Mauritania.
| |  |
|
|
| In this second, we see an expression of the uncertainty in the model, calculated as the difference between the belief and plausibility images.
| |  |
|
|
| Images may be transformed into measures of fuzzy set membership with the module FUZZY. Three common function shapes are provided: linear, S-shaped, and J-shaped. User-defined functions are also supported. This technique is used, for example, to standardize the multiple layers of a multi-criteria suitability analysis prior to aggregation. Images in disparate units of measurement are all standardized to the same scale of suitability and may then be combined to produce a single aggregate suitability image.
| |  |
|
;)
;)
;)
;)
;)
;)
