Results Analytics scores and ranks each alternative based on a combination of the thresholds and objective assigned to the parameter along with the weight assigned to the parameter and the ranking method. Alternatives in the data table are sorted by rank, which is inversely related to the score—the alternative with the highest score is given a rank of one.
You can group parameters, and the scoring of parameter groups is hierarchical—the
effects of priorities are cascaded from parent groups to
child groups and down to parameters. All requirements are rolled up to
determine an overall score for an alternative.
Note:
If the data table is displaying both the infeasible data points and the best designs, the infeasible data points may be scored higher than the best designs. Data points that violate a threshold are not penalized for this violation unless the parameter has a MH (Must Have) priority. As a result, infeasible data points that violate thresholds for a parameter with P1 to P5 priority can have higher scores than best designs with no threshold violations.
Requirements
The user-defined requirements for an alternative are its parameter grouping, priority, thresholds (upper and lower values), and objective. You can choose to maximize the objective, minimize the objective, or enter a target value for the objective.
Final Weight Factor
Results Analytics uses two interdependent variables to control the final weight factor of any parameter or group of parameters—priority and initial weight. However, if you subsequently change an initial weight on the
Collaborate page, the weight that you assign overrides the weight calculated by
Results Analytics from the assigned priority.
- Priority
-
A priority specifies the relative importance of a parameter and its associated constraint or objective, where a priority of 1 is higher than a priority of 5 (the default value). You assign priorities without prior knowledge of their effect on the overall scoring and ranking and without information about their effect on tradeoffs. Results Analytics uses the priority value to initialize the weight, which is then used for scoring and ranking the alternatives. For example, in the car comparison analytic case, you might decide that maximizing the horsepower of the vehicle is more important than minimizing the price and assign priorities of 1 and 3, respectively.
The following table shows how Results Analytics uses the priority to initialize the weight:
| Priority | Weight |
|---|
| MH | 5.0 |
| P1 | 5.0 |
| P2 | 4.0 |
| P3 | 3.0 |
| P4 | 2.0 |
| P5 | 1.0 |
- Initial Weight
- A weight is similar to a priority in that it specifies the relative importance of a parameter, where a weight of 10 is higher than a weight of 1. However, unlike priorities, you adjust the weights after applying the initial scoring and ranking and after filtering out data points; for example, after removing data points that fail to meet threshold values. You (and your collaborators) can then adjust the weights to gain a better understanding of the effects of tradeoffs and how the parameters combine to influence the scoring and ranking. For example, relaxing the weight of one parameter, say the cost of the vehicle, may significantly increase the influence of a second parameter, say the fuel consumption. After you have selected alternatives, you can use the Collaborate page to perform weight trade-off analyses. You can assign a weight between 1 and 5 to a parameter, where 1 (default) is the lowest weight and 5 the highest weight. When you assign a weight to a parameter, the assigned weight overrides the weight calculated by Results Analytics from the assigned priority.
Weights and Groups
If some of the parameters in your data set are grouped, the selected ranking
method uses the groups to calculate the final weight factor:
- Results Analytics adds all
of the weights
for a child group and its parameters to
get a sum of weights within a given parent group.
Results Analytics divides each
weight
for a child group or a parameter by the
above sum to get a fraction of the total sum. This value is
the
fractional weight
of a child group or a parameter.
- The final weight factor of a child group or a parameter is the product of its
own
fractional weight and the
fractional weights
of all parent groups above this group or
parameter.