The invention relates to a method of multiple criteria decision making. In particular, a method of maintaining a complex article is described that includes the steps of; (a) taking a set of options for replacing each of a plurality of components of the complex article, (b) identifying an optimum solution set from the set of possible replacement options of step (a) that best satisfies a plurality of criteria and (c) implementing an optimum solution of the optimum solution set. The step of identifying an optimum solution set comprises the step of evaluating a plurality of potential solutions from the set of possible options defined in step (a) against the plurality of criteria. The criteria values for potential solutions comprise probability functions and the step of evaluating a potential solution comprises the step of calculating a measure of effectiveness (MOE) of the potential solution from the probability function of each criterion value and at least one associated user function defining the acceptability of said criterion. The use of optimisation processes, such as evolutionary algorithms (including genetic algorithms) and dynamic programming are also disclosed. The application of the process to various alternative problems is also outlined.