Hybrid Value-Utility Decision Analysis

Abstract

The Army faces complex decisions containing many variables and sources of uncertainty. Under standard DA approaches, value functions are employed under conditions of certainty and utility functions under conditions of. uncertainty. Utility functions are unique to an individual, and so must be elicited from the decision maker. The time of military decision makers is extremely valuable, and makes employment of the utility approach problematic. The concept of a hybrid value-utility DA model is promulgated. Such a model may be constructed beginning with a value model based on elicitations from subject matter experts, e.g., staff officers. Selected single dimensional value functions are replaced with the corresponding utility functions to form a hybrid model. Substitutions are prioritized through response surface methodology applied to the value functions of the decision model. Substitutions continue until the performance of the hybrid suitably estimates the performance of the true utility model. Construction of such a hybrid model minimizes the burden placed on the decision maker while maintaining the benefits of DA. A sample case illustrates the elicitation efficiency where an automatic target recognition classification system choice attribute set was reduced from 23 to eight.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2002
Accession Number
ADA403768

Entities

People

  • William K. Klimack

Organizations

  • United States Military Academy

Tags

Communities of Interest

  • C4I
  • Human Systems

DTIC Thesaurus Topics

  • Air Force
  • Algorithms
  • Classification
  • Computational Science
  • Detection
  • Employment
  • Engineering
  • Experimental Design
  • Identification
  • Money
  • Operations Research
  • Recognition
  • Standards
  • Systems Engineering
  • Target Recognition
  • United States
  • United States Military Academy

Readers

  • Computational Modeling and Simulation
  • Joint Military Operations and Doctrine.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.