Introduction
Swing weight is a quantitative measure that represents the relative importance of an objective in decision analysis, calculated based on stakeholder preferences and importance rankings.
Overview
Swing weight is a key component of the ISEDM (Integrated Systems Engineering Decision Management) methodology. It quantifies the relative importance of different objectives in a decision-making process, allowing for a weighted sum of individual objective values to determine an overall performance score. The weight is determined through a constrained optimization process based on SME (Subject Matter Expert) rankings of importance and differentiation.
Swing weights directly impact the calculation of the overall performance value for each alternative. The weights are constrained to sum to 1, ensuring the overall value for any alternative is between 0 and 100.
| Swing weights are recalculated every time the importance or differentiation settings change in the Objectives Tab, and changes must be committed using the "Submit" button. |
Position in Knowledge Hierarchy
Broader concepts: - ISEDM (is-a)
Details
Swing weights are calculated using a constrained optimization problem based on SME and stakeholder rankings of the importance and differentiation of the objectives. The calculation process involves:
-
Ranking objectives based on importance (Defining to Enabling) and differentiation (Low to High)
-
Flattening these rankings into a single position in an ascending ranking of weight value
-
Formulating a sparse matrix of inequality constraints
-
Solving the constrained optimization problem
The relationship between importance, differentiation, and swing weight is illustrated in Figure 125 of the context, which shows how SME rankings are translated into weight values.
| All classifications (and thus swing weights) are relative to one another. Settings directly impact swing weights. Swing weights and values recalculate every time the settings change. |
| Swing weights are not static values. They change whenever the importance or differentiation settings are modified, and the changes must be committed using the "Submit" button in the dashboard. |
The following table shows how importance and differentiation rankings translate to swing weights:
Objective |
Importance Ranking |
Differentiation Ranking |
Swing Weight |
Cost |
3 |
2 |
0.4 |
Performance |
1 |
3 |
0.2 |
Time |
2 |
1 |
0.4 |
This example shows how the rankings (where lower numbers indicate higher importance) translate to swing weights that sum to 1. The calculation process ensures that more important and more differentiated objectives receive higher weights.
Practical applications and examples
Swing weights are used in the ISEDM methodology to calculate the overall performance value for each alternative. The process is as follows:
-
For each alternative (i), for every Measure of Effectiveness (MOE) (j):
-
Interpolate the MOE measure to a value measure using the single attribute value function
-
Weight that value using the swing weight for that objective (0-1)
-
Add to the running sum of performance value for the alternative
-
-
The overall value is the weighted sum of individual objective values.
The process is illustrated in Figure 123 and Figure 124 of the context, showing how swing weights are applied to value functions to calculate the overall performance score.
| The Decision Analysis Value Function tab (Figure 123) allows for adjusting the value functions, and the Swing Weight tab (Figure 125) shows how the weights are calculated based on the importance and differentiation settings. |
| Only active objectives (those checked in the Objectives Tab) are used to compute swing weights and performance values. Inactive objectives do not contribute to the overall score calculation. |
Related wiki pages
References
Knowledge Graph
Associated Diagrams
