Decisions with Multiple Objectives

Abstract

Many of the complex problems faced by decision makers involve multiple conflicting objectives. This book describes how a confused decision maker, who wishes to make a reasonable and responsible choice among alternatives, can systematically probe his true feelings in order to make those critically important, vexing trade-offs between incommensurable objectives. The theory is illustrated by many real concrete examples taken from a host of disciplinary settings. The standard approach in decision theory or decision analysis specifies a simplified single objective like monetary return to maximise. By generalising from the single objective case to the multiple objective case, this book considerably widens the range of applicability of decision analysis.

Predicting adverse impact and mean criterion performance in multistage selection.

“…In the language of multicriteria optimization, these optimal trade-off scenarios are also called Pareto optimal in the sense that any other feasible scenario that differs from them results in either a decrease of the selection quality or an increase of the cost or the level of adverse impact, and the entire set of optimal scenarios (and, in particular, the set of associated trade-offs) is usually referred to as the Pareto surface (Keeney & Raiffa, 1993). Given a tabular or, preferably, a pictorial representation of this Pareto surface, the practitioner could then decide on the preferred optimal trade-off and the associated scenario.…”

Construction of a Choquet integral and the value functions without any commensurateness assumption in multi-criteria decision making

“…There is in [20] (see also [21]) a major result regarding the construction of value functions in conjoint measurement. The aim is to construct a measurable value function v : X → R from preference order relations.…”

Age Effects and Heuristics in Decision Making

“…It is measured as the percentage of states covered by the option. Third is the lexicographic heuristic which favors options that cover the most probable state (Keeney andRaiffa 1993, Gigerenzer andGoldstein 1996). If this does not lead to a unique choice, the second most probable state is used, and so on.…”