A Pictorial Guide for the Fuzzy AHP Online Software
Fuzzy AHP (Chang) Software is easy to use: (1) draw the hierarchical chart, (2) select/create your own fuzzy scale, (3) fill the pairwise comparison matrices, and (4) get the full output.
1. Enter the project name
First of all, enter your project’s name.
For example, in this project, the fuzzy AHP (Chang) method is used to select the best school.
2.Design a hierarchical chart
Here, you can create your own hierarchical chart
To add a criterion or sub-criterion, click the plus (+) symbol, and to delete a criterion or sub-criterion, click the minus (-) symbol.
To write the desired name for the criterion, click on the criterion name and enter your own desired name that is short to write.
As shown on the picture, to set the chart’s appearance, click the gear icon to change the color, shape and any other settings (this feature will be added soon)
If there are several alternatives, select “Yes” and then click “Add Alternative” button to add the new alternative. If there is no alternative, select “No”.
To delete a selected alternative, click the minus (-) symbol.
Changing the name and the appearance of alternative is exactly the same way used for the criterion (this feature will be added soon).
3. Select /create the fuzzy scale
Here, you can select or create your own fuzzy scale from a list of default fuzzy scales.
You can select your own default fuzzy scale from a list of scales in the slider bar.
If you tend to create your own scale, click “Create your Scale” and create your own scale in the table.
To create a new scale, you need to enter your own verbal expressions and fuzzy triangular numbers. A triangular fuzzy number can be denoted as ( l , m , u ) , where l and u stand for the lower and upper values of fuzzy numbers, respectively, and m for the middle value.
As shown on the picture, you can easily edit a scale
4. Enter pairwise comparisons
Here, you can fill the pairwise comparison matrices created based on the hierarchical chart.
Notice that in the pairwise comparison matrices, the codes that appear in the scale table need to enter. (A code represents a verbal expression, for example, code 1 implies equal importance)
To fill the pairwise comparison tables, simply click in each cell of the matrix and enter your own pairwise comparison number.
It is recommended that you insert codes in Excel in order to avoid wasting time, and then copy and paste tables from Excel at once.
If a row criterion is more important than the column criterion, enter a positive number. For example, if you enter the 5, the row criterion is 5 times more important than the column criterion.
If a column criterion is more important than the row criterion, enter a negative number. For example, if you enter – 4, the column criterion is 4 times more important than the row criterion.
By clicking on the “Inconsistency Ratio” button, the inconsistency ratio can be displayed at the top of each pairwise comparison matrix.
As you can see from the picture above, experts include the following:
Add Expert: If there several experts, you can increase the number of experts by clicking “Add Expert”. By clicking on this button, a previous expert will be automatically saved and the new expert will be added.
Delete Expert: By clicking “Delete Expert”, the current expert will be deleted.
Save Expert: By clicking “Save Expert”, the current expert data will be saved.
As you can see in the picture below, the status of experts include the following:
If all the expert data are completed and stored, the expert’s status is “completed”.
If all the expert data are not completed, the expert’s status is “incompleted”.
If an expert has been deleted, the expert’s status is “deleted” and you can retrieve the data.
mean: By clicking on the “mean” button, the fuzzy average of all the experts’ judgments along with the inconsistency ratio for all of the pairwise comparison matrices will be observed. (How to calculate the meanis described by a link in the existing page)
To observe the average, the data of all current experts needs to be completed.
There a scale bar at the bottom of the screen that displays the preference value of the selected cell and you can change the pairwise comparison number using the scale bar tab.
If the entered number is positive, the row component is preferred over the column component.