1. ¹Intelligence Systems, Great Malvern, UK
2. ²PCS Department, Farnborough, UK
3. ³University of Warwick, Coventry, UK

Correspondence: L Dodd, Intelligence Systems, QinetiQ, Room 208, Turing Building, St Andrews Road, Great Malvern, Worcs WR14 3PS, UK. E-mail: LDODD@qinetiq.com

Received 0 October 2004; Accepted 0 May 2005; Published online 24 August 2005.

Abstract

In previous work, we considered the representation of human decision-making processes in closed-form simulation models of conflict. An important element of this representation is the rapid planning process that embodies the processing of information for situation assessment to support a course of action decision (eg in a military headquarters). The application of this work is in support of operational analysis models for defence procurement and balance of investment. This paper describes the application of non-linear multi-attribute utility theory in conflict scenarios in order to extend the representation of the rapid planning process to account for a wider set of subjective attributes of the decision-maker. The results show, through examination of experimental data, that decision-making can be modelled through a particular class of utility functions. These utilities embody a geometry which allows us to classify the types of decision being made when there are conflicting objectives and when decision-makers adopt very different and subjective appraisals of constraints and beliefs in outcome. The experimental results help to demonstrate that the subjective nature of the situation assessment, and the personality, training, experience and history of the decision-maker are central to the functional representations. This paper presents a way to capture this deeper representation of human decision-making in a way that is potentially useful for quantitative modelling using the rapid planning process as a basis.