1Robustness Against the Decision-Maker’s Attitude to Risk in Problems With Conflicting Objectives.pdf

IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, VOL. 16, NO. 1, FEBRUARY 2012 1 Robustness Against the Decision-Maker’s Attitude to Risk in Problems With Conflicting Objectives Lam T. Bui, Hussein A. Abbass, Senior Member, IEEE, Michael Barlow, and Axel Bender Abstract —In multiobjective optimization problems (MOPs), I. Introduction the Pareto set consists of efficient solutions that represent the best trade-offs between the conflicting objectives. Many forms of EAL-WORLD optimization problems are characterized uncertainty affect the MOP, including uncertainty in the decision Rby conflicting objectives and high levels of uncertainty. variables, parameters or objectives. A source of uncertainty that Those who will use the optimization result—e.g., want to make is not studied in the evolutionary multiobjective optimization the best possible decision given the optimization outputs—are (EMO) literature is the decision-maker’s attitude to risk (DMAR) faced with a situation where they need to trade off benefits even though it has great significance in real-world applications. Often the decision-makers change over the course of the decision- and drawbacks amongst alternatives that are hard to compare. making process and thus, some relevant information about Picking a solution thus becomes a matter of preference . Given preferences of future decision-makers is unknown at the time a the high levels of uncertainty, this is significantly influenced decision is made. This poses a major risk to organizations because by attitudes to risk with “risk” here being defined as the a new decision-maker may simply