cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Submit your abstract to the call for content for Discovery Summit Americas by April 23. Selected abstracts will be presented at Discovery Summit, Oct. 21- 24.
Discovery is online this week, April 16 and 18. Join us for these exciting interactive sessions.
Choose Language Hide Translation Bar
Trade-Off Plot Add-In
seburke89
Level III

The Trade-Off Plot Add-In for JMP allows users to compare the tradeoffs of multiple criteria to identify promising candidate solutions that provide the desired balance for a particular set of objectives. Each row in the data table represents a potential solution under consideration. Columns represent the numerical scores for the various objectives/criteria under consideration. After installing, select the Add-In with the data table open.


The first step is to identify the columns to be considered as criteria and specify whether they are optimized by maximizing or minimizing. If the goal is to hit a target value, a constructed column of the absolute value of the criterion minus the target value can be constructed. The Add-In then identifies the solutions on the Pareto front to objectively eliminate inferior solutions from further consideration. A Pareto front allows examination of the tradeoffs of multiple objectives or criteria in a given scenario. A solution is on the Pareto front if it is not dominated by any other point in the obtainable criterion region. The Add-In creates a new data table that contains only the solutions on the Pareto front based on the supplied information. All other solutions can objectively be removed from consideration as not rational choices.


The results window has four components:


The first is a numeric summary table of the criteria – the best and worst values across the Pareto Front and original data set are shown.


There are three graphs to visualize results. The first two are scatterplot matrices of the criteria under consideration, graphing just the solutions on the Pareto front. The first scatterplot matrix shows the criteria in their original units. The second scatterplot matrix shows the criteria scaled to desirability values between 0 and 1 (0 being worst and 1 being best). There are two options to scale the criteria on a [0,1] scale: based on the worst values in the original data set or based on the worst values in the Pareto front.

 

The final graph is the trade-off plot to allow the user to visually examine the trade-offs across multiple criteria. The solutions are sorted from left to right from worst to best for one of the criteria. Each other criterion is shown with a separate line. On the left-hand side of the window, the user can change which criterion is sorted from worst to best. All the plots are dynamic and connected with the data table of the Pareto front solutions. If the user selects a point in any plot, those points will be highlighted in the other plots and data table.