Online Program

Saturday, February 22
CS24 Using Graphs in Decisionmaking and Quality Control Sat, Feb 22, 10:45 AM - 12:15 PM
Bayshore VII

Improved Decisionmaking When Balancing Multiple Objectives (302697)

*Christine Anderson-Cook, Los Alamos National Laboratory 
Lu Lu, University of South Florida 

Keywords: trade-offs, Pareto front, multi-faceted optimization

When we are faced with choosing the best solution for a problem, the right solution is obvious if there’s only a single objective. However, in practice, decisions are rarely that simple. Often, there are several competing objectives that must be balanced, and team members may value objectives differently. This talk will provide a formal approach for improved decisionmaking in these situations. We consider the Pareto front approach, which begins with an objective phase that eliminates all nonrational candidates from further consideration based on their quantitative criterion values. In the second subjective phase, the practitioners’ priorities of the decisions are incorporated through the use of graphical tools that allow comparison of the remaining alternatives based on their relative performance. The tools also explore which desirability function weights different solutions are best for and how the solutions compare to the best available choice for a given set of weights. The methods are explained and illustrated with examples with different numbers of competing objectives. Available software developed by the authors will be used to demonstrate the methods.