You can use The Concept System Core Program as the basis for making complex multifaceted decisions.

For instance, let’s assume your organization has to decide among one of three alternative plans to address some issue. You might generate statements that describe the features that would be desirable in an “ideal” plan. The participants would sort and rate (for importance or desirability) the feature statements. Then, you would set up one rating for each of the three options you’re considering. For each option, you would rate how well the option addresses each of the features. This is sometimes referred to as a utility rating. When you’ve developed and saved a map, you can then use pattern matching to compare the three options and decide which is considered “best” by different stakeholder groups. You would do this using the “Utility Score” function of the pattern matching module. For each option, you would obtain an overall utility by combining the feature importance rating (on the left axis) with the degree to which that option addresses each feature (on the right axis). The option with the highest overall utility score is considered the “best” choice for those conditions.

What is the Utility Score?

The Utility Score in a pattern matching ladder graph is simply the sum of the cross-products of the variables on the left and right axes. Or, to get the Utility Score, for each cluster we multiply the average importance by the average utility rating, and then we sum this product across clusters. For instance, if Option A is a good option, it should have a high utility rating on clusters that were rated as high in importance, and lower utility ratings for clusters rated lower in importance. The cross-product of these two variables should be fairly high. On the other hand, if Option B is a poor option, it will have low utility ratings for clusters that are high in importance, and the cross-product of these two variables will be lower. The Utility Score -- the sum of these cross-products -- is our best estimate of which option overall has the greatest utility on the most important features. In some contexts it is desirable to ‘normalize’ one or both of the variables before doing this cross-multiplication (a normalized variable is a variable whose values sum to one).

This is essentially a simple form of a method that is a branch of decision analysis called Multi-Attribute Utility Theory, or MAUT for short. For a nice introduction to MAUT, you might look at:

Edwards, W. and Newman, J.R. (1982). Multiattribute Evaluation. Sage Publications, Newbury Park, California.